chore(main): release 0.7.0 (#817)

Co-authored-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com>

.env.example

@@ -145,9 +145,44 @@ ANTHROPIC_API_KEY=sk-ant-your-key-here
 # CLAUDE_CODE_USE_OPENAI=1
 # OPENAI_API_KEY=sk-your-key-here
 # OPENAI_MODEL=gpt-4o
+# For DeepSeek, set:
+# OPENAI_BASE_URL=https://api.deepseek.com/v1
+# OPENAI_MODEL=deepseek-v4-flash
+# Optional: OPENAI_MODEL=deepseek-v4-pro
+# Legacy aliases also work: deepseek-chat and deepseek-reasoner
+# For Z.AI GLM Coding Plan, set:
+# OPENAI_BASE_URL=https://api.z.ai/api/coding/paas/v4
+# OPENAI_MODEL=GLM-5.1
+# Optional: OPENAI_MODEL=GLM-5-Turbo, GLM-4.7, or GLM-4.5-Air
 
 # Use a custom OpenAI-compatible endpoint (optional — defaults to api.openai.com)
 # OPENAI_BASE_URL=https://api.openai.com/v1
+# Choose the OpenAI-compatible API surface (optional — defaults to chat_completions)
+# Supported: chat_completions, responses
+# OPENAI_API_FORMAT=chat_completions
+# Choose a custom auth header for OpenAI-compatible providers (optional).
+# Authorization defaults to Bearer; custom headers default to the raw API key.
+# Set OPENAI_AUTH_HEADER_VALUE when the header value differs from OPENAI_API_KEY.
+# OPENAI_AUTH_HEADER=api-key
+# OPENAI_AUTH_SCHEME=raw
+# OPENAI_AUTH_HEADER_VALUE=your-header-value-here
+
+# Fallback context window size (tokens) when the model is not found in the
+# built-in table (default: 128000). Increase this for models with larger
+# context windows (e.g. 200000 for Claude-sized contexts).
+# CLAUDE_CODE_OPENAI_FALLBACK_CONTEXT_WINDOW=128000
+
+# Per-model context window overrides as a JSON object.
+# Takes precedence over the built-in table, so you can register new or
+# custom models without patching source.
+# Example: CLAUDE_CODE_OPENAI_CONTEXT_WINDOWS={"my-corp/llm-v3":262144,"gpt-4o-mini":128000}
+# CLAUDE_CODE_OPENAI_CONTEXT_WINDOWS=
+
+# Per-model maximum output token overrides as a JSON object.
+# Use this alongside CLAUDE_CODE_OPENAI_CONTEXT_WINDOWS when your model
+# supports a different output limit than what the built-in table specifies.
+# Example: CLAUDE_CODE_OPENAI_MAX_OUTPUT_TOKENS={"my-corp/llm-v3":8192}
+# CLAUDE_CODE_OPENAI_MAX_OUTPUT_TOKENS=
 
 
 # -----------------------------------------------------------------------------
@@ -267,6 +302,30 @@ ANTHROPIC_API_KEY=sk-ant-your-key-here
 # Disable "Co-authored-by" line in git commits made by OpenClaude
 # OPENCLAUDE_DISABLE_CO_AUTHORED_BY=1
 
+# Disable strict tool schema normalization for non-Gemini providers
+# Useful when MCP tools with complex optional params (e.g. list[dict])
+# trigger "Extra required key ... supplied" errors from OpenAI-compatible endpoints
+# OPENCLAUDE_DISABLE_STRICT_TOOLS=1
+
+# Disable hidden <system-reminder> messages injected into tool output
+# Suppresses the file-read cyber-risk reminder and the todo/task tool nudges
+# Useful for users who want full transparency over what the model sees
+# OPENCLAUDE_DISABLE_TOOL_REMINDERS=1
+
+# Log structured per-request token usage (including cache metrics) to stderr.
+# Useful for auditing cache hit rate / debugging cost spikes outside the REPL.
+# Any truthy value enables it ("verbose", "1", "true").
+#
+# Complements (does NOT replace) CLAUDE_CODE_ENABLE_TOKEN_USAGE_ATTACHMENT —
+# they serve different audiences:
+#   - OPENCLAUDE_LOG_TOKEN_USAGE is user-facing: one JSON line per API
+#     request on stderr, intended for humans inspecting cost/caching.
+#   - CLAUDE_CODE_ENABLE_TOKEN_USAGE_ATTACHMENT is model-facing: injects
+#     a context-usage attachment INTO the prompt so the model can reason
+#     about its own remaining context. Does not touch stderr.
+# Turn on whichever audience you're debugging; both can run together.
+# OPENCLAUDE_LOG_TOKEN_USAGE=verbose
+
 # Custom timeout for API requests in milliseconds (default: varies)
 # API_TIMEOUT_MS=60000
 

.gitignore

@@ -11,3 +11,4 @@ CLAUDE.md
 package-lock.json
 /.claude
 coverage/
+agent.log

.release-please-manifest.json

@@ -1,3 +1,3 @@
 {
-  ".": "0.6.0"
+  ".": "0.7.0"
 }

CHANGELOG.md

@@ -1,5 +1,39 @@
 # Changelog
 
+## [0.7.0](https://github.com/Gitlawb/openclaude/compare/v0.6.0...v0.7.0) (2026-04-26)
+
+
+### Features
+
+* add model-specific tokenizers and compression ratio detection ([#799](https://github.com/Gitlawb/openclaude/issues/799)) ([e92e527](https://github.com/Gitlawb/openclaude/commit/e92e5274b223d935d380b1fbd234cb631ab03211))
+* add OPENCLAUDE_DISABLE_TOOL_REMINDERS env var to suppress hidden tool-output reminders ([#837](https://github.com/Gitlawb/openclaude/issues/837)) ([28de94d](https://github.com/Gitlawb/openclaude/commit/28de94df5dcd7718cb334e2e793e9472f5b291c5)), closes [#809](https://github.com/Gitlawb/openclaude/issues/809)
+* add streaming optimizer and structured request logging ([#703](https://github.com/Gitlawb/openclaude/issues/703)) ([5b9cd21](https://github.com/Gitlawb/openclaude/commit/5b9cd21e373823a77fd552d6e02f5d4b68ae06b1))
+* add xAI as official provider ([#865](https://github.com/Gitlawb/openclaude/issues/865)) ([2586a9c](https://github.com/Gitlawb/openclaude/commit/2586a9cddbd2512826bca81cb5deb3ec97f00f0f))
+* **api:** expose cache metrics in REPL + normalize across providers ([#813](https://github.com/Gitlawb/openclaude/issues/813)) ([9e23c2b](https://github.com/Gitlawb/openclaude/commit/9e23c2bec43697187762601db5b1585c9b0fb1a3))
+* implement Hook Chains runtime integration for self-healing agent mesh MVP ([#711](https://github.com/Gitlawb/openclaude/issues/711)) ([44a2c30](https://github.com/Gitlawb/openclaude/commit/44a2c30d5f9b98027e454466c680360f6b4625fc))
+* **memory:** implement persistent project-level Knowledge Graph and RAG ([#899](https://github.com/Gitlawb/openclaude/issues/899)) ([29f7579](https://github.com/Gitlawb/openclaude/commit/29f757937732be0f8cca2bc0627a27eeafc2a992))
+* **minimax:** add /usage support and fix MiniMax quota parsing ([#869](https://github.com/Gitlawb/openclaude/issues/869)) ([26413f6](https://github.com/Gitlawb/openclaude/commit/26413f6d307928a4f14c9c61c9860a28f8d81358))
+* **model:** add GPT-5.5 support for Codex provider ([#880](https://github.com/Gitlawb/openclaude/issues/880)) ([038f715](https://github.com/Gitlawb/openclaude/commit/038f715b7ab9714340bda421b73a86d8590cf531))
+* **tools:** resilient web search and fetch across all providers ([#836](https://github.com/Gitlawb/openclaude/issues/836)) ([531e3f1](https://github.com/Gitlawb/openclaude/commit/531e3f10592a73d81f26675c2479d46a3d5b55f5))
+* **zai:** add Z.AI GLM Coding Plan provider preset ([#896](https://github.com/Gitlawb/openclaude/issues/896)) ([a0d657e](https://github.com/Gitlawb/openclaude/commit/a0d657ee188f52f8a4ceaad1658c81343a32fdad))
+
+
+### Bug Fixes
+
+* **agent:** provider-aware fallback for haiku/sonnet aliases ([#908](https://github.com/Gitlawb/openclaude/issues/908)) ([a3e728a](https://github.com/Gitlawb/openclaude/commit/a3e728a114f6379b80daefc8abcac17a752c5f96))
+* bugs ([#885](https://github.com/Gitlawb/openclaude/issues/885)) ([c6c5f06](https://github.com/Gitlawb/openclaude/commit/c6c5f0608cf6509b412b121954547d72b3f3a411))
+* make OpenAI fallback context window configurable + support external model lookup ([#861](https://github.com/Gitlawb/openclaude/issues/861)) ([b750e9e](https://github.com/Gitlawb/openclaude/commit/b750e9e97d15926d094d435772b2d6d12e5e545c))
+* **mcp:** disable MCP_SKILLS feature flag — source not mirrored ([#872](https://github.com/Gitlawb/openclaude/issues/872)) ([dcbe295](https://github.com/Gitlawb/openclaude/commit/dcbe29558ab9c74d335b138488005a6509aa906a))
+* normalize /provider multi-model selection and semicolon parsing ([#841](https://github.com/Gitlawb/openclaude/issues/841)) ([c4cb98a](https://github.com/Gitlawb/openclaude/commit/c4cb98a4f092062da02a4728cf59fed0fc3a6d3f))
+* **openai-shim:** echo reasoning_content on assistant tool-call messages for Moonshot ([#828](https://github.com/Gitlawb/openclaude/issues/828)) ([67de6bd](https://github.com/Gitlawb/openclaude/commit/67de6bd2cffc3381f0f28fd3ffce043970611667))
+* **query:** restore system prompt structure and add missing config import ([#907](https://github.com/Gitlawb/openclaude/issues/907)) ([818689b](https://github.com/Gitlawb/openclaude/commit/818689b2ee71cb6966cb4dc5a5ebd90fd22b0fcb))
+* **shell:** recover when CWD path was replaced by a non-directory ([#871](https://github.com/Gitlawb/openclaude/issues/871)) ([a4c6757](https://github.com/Gitlawb/openclaude/commit/a4c67570238794317d049a225396672b465fdbfc))
+* **startup:** show --model flag override on startup screen ([#898](https://github.com/Gitlawb/openclaude/issues/898)) ([d45628c](https://github.com/Gitlawb/openclaude/commit/d45628c41300b83b466e6a97983099615a50e7d7))
+* **startup:** url authoritative over model name in banner provider detect ([#864](https://github.com/Gitlawb/openclaude/issues/864)) ([e346b8d](https://github.com/Gitlawb/openclaude/commit/e346b8d5ec2d58a4e8db337918d52d844ee52766)), closes [#855](https://github.com/Gitlawb/openclaude/issues/855)
+* surface actionable error when DuckDuckGo web search is rate-limited ([#834](https://github.com/Gitlawb/openclaude/issues/834)) ([3c4d843](https://github.com/Gitlawb/openclaude/commit/3c4d8435c42e1ee04f9defd31c4c589017f524c5))
+* **test:** add missing teammate exports to hookChains integration mock ([#840](https://github.com/Gitlawb/openclaude/issues/840)) ([23e8cfb](https://github.com/Gitlawb/openclaude/commit/23e8cfbd5b22179684276bef4131e26b830ce69c)), closes [#839](https://github.com/Gitlawb/openclaude/issues/839)
+* **update:** show real package version and give actionable guidance ([#870](https://github.com/Gitlawb/openclaude/issues/870)) ([6e58b81](https://github.com/Gitlawb/openclaude/commit/6e58b819370128b923dda4fcc774bb556f4b951a))
+
 ## [0.6.0](https://github.com/Gitlawb/openclaude/compare/v0.5.2...v0.6.0) (2026-04-22)
 
 

PLAYBOOK.md

@@ -132,7 +132,7 @@ Cause:
 Fix:
 
 ```powershell
-cd C:\Users\Lucas Pedry\Documents\openclaude\openclaude
+cd <PATH>
 bun run dev:profile
 ```
 
@@ -189,7 +189,7 @@ Or pick a local Ollama profile automatically by goal:
 bun run profile:init -- --provider ollama --goal balanced
 ```
 
-## 6.5 Placeholder key (`SUA_CHAVE`) error
+## 6.5 Placeholder key (`YOUR_KEY`) error
 
 Cause:
 

README.md

@@ -13,7 +13,25 @@ Use OpenAI-compatible APIs, Gemini, GitHub Models, Codex OAuth, Codex, Ollama, A
 OpenClaude is also mirrored to GitLawb:
 [gitlawb.com/node/repos/z6MkqDnb/openclaude](https://gitlawb.com/node/repos/z6MkqDnb/openclaude)
 
-[Quick Start](#quick-start) | [Setup Guides](#setup-guides) | [Providers](#supported-providers) | [Source Build](#source-build-and-local-development) | [VS Code Extension](#vs-code-extension) | [Community](#community)
+[Quick Start](#quick-start) | [Setup Guides](#setup-guides) | [Providers](#supported-providers) | [Source Build](#source-build-and-local-development) | [VS Code Extension](#vs-code-extension) | [Sponsors](#sponsors) | [Community](#community)
+
+## Sponsors
+
+<p align="center">
+  <a href="https://gitlawb.com">
+    <img src="https://gitlawb.com/logo.png" alt="GitLawb logo" width="96">
+  </a>
+  &nbsp;&nbsp;&nbsp;&nbsp;
+  <a href="https://bankr.bot">
+    <img src="https://bankr.bot/favicon.svg" alt="Bankr.bot logo" width="96">
+  </a>
+</p>
+
+<p align="center">
+  <a href="https://gitlawb.com"><strong>GitLawb</strong></a>
+  &nbsp;&nbsp;&nbsp;&nbsp;
+  <a href="https://bankr.bot"><strong>Bankr.bot</strong></a>
+</p>
 
 ## Star History
 
@@ -157,7 +175,7 @@ Add to `~/.claude/settings.json`:
 ```json
 {
   "agentModels": {
-    "deepseek-chat": {
+    "deepseek-v4-flash": {
       "base_url": "https://api.deepseek.com/v1",
       "api_key": "sk-your-key"
     },
@@ -167,10 +185,10 @@ Add to `~/.claude/settings.json`:
     }
   },
   "agentRouting": {
-    "Explore": "deepseek-chat",
+    "Explore": "deepseek-v4-flash",
     "Plan": "gpt-4o",
     "general-purpose": "gpt-4o",
-    "frontend-dev": "deepseek-chat",
+    "frontend-dev": "deepseek-v4-flash",
     "default": "gpt-4o"
   }
 }

docs/advanced-setup.md

@@ -68,9 +68,11 @@ openclaude
 export CLAUDE_CODE_USE_OPENAI=1
 export OPENAI_API_KEY=sk-...
 export OPENAI_BASE_URL=https://api.deepseek.com/v1
-export OPENAI_MODEL=deepseek-chat
+export OPENAI_MODEL=deepseek-v4-flash
 ```
 
+Use `deepseek-v4-pro` when you want the stronger model. `deepseek-chat` and `deepseek-reasoner` remain available as DeepSeek's legacy API aliases.
+
 ### Google Gemini via OpenRouter
 
 ```bash
@@ -169,12 +171,13 @@ export OPENAI_MODEL=gpt-4o
 |----------|----------|-------------|
 | `CLAUDE_CODE_USE_OPENAI` | Yes | Set to `1` to enable the OpenAI provider |
 | `OPENAI_API_KEY` | Yes* | Your API key (`*` not needed for local models like Ollama or Atomic Chat) |
-| `OPENAI_MODEL` | Yes | Model name such as `gpt-4o`, `deepseek-chat`, or `llama3.3:70b` |
+| `OPENAI_MODEL` | Yes | Model name such as `gpt-4o`, `deepseek-v4-flash`, or `llama3.3:70b` |
 | `OPENAI_BASE_URL` | No | API endpoint, defaulting to `https://api.openai.com/v1` |
 | `CODEX_API_KEY` | Codex only | Codex or ChatGPT access token override |
 | `CODEX_AUTH_JSON_PATH` | Codex only | Path to a Codex CLI `auth.json` file |
 | `CODEX_HOME` | Codex only | Alternative Codex home directory |
 | `OPENCLAUDE_DISABLE_CO_AUTHORED_BY` | No | Suppress the default `Co-Authored-By` trailer in generated git commits |
+| `OPENCLAUDE_LOG_TOKEN_USAGE` | No | When truthy (e.g. `verbose`), emits one JSON line on stderr per API request with input/output/cache tokens and the resolved provider. **User-facing debug output** — complements the REPL display controlled by `/config showCacheStats`. Distinct from `CLAUDE_CODE_ENABLE_TOKEN_USAGE_ATTACHMENT`, which is **model-facing** (injects context usage info into the prompt itself). Both can run together. |
 
 You can also use `ANTHROPIC_MODEL` to override the model name. `OPENAI_MODEL` takes priority.
 

docs/hook-chains.md

@@ -0,0 +1,333 @@
+# Hook Chains (Self-Healing Agent Mesh MVP)
+
+Hook Chains provide an event-driven recovery layer for important workflow failures.
+When a matching hook event occurs, OpenClaude evaluates declarative rules and can dispatch remediation actions such as:
+
+- `spawn_fallback_agent`
+- `notify_team`
+- `warm_remote_capacity`
+
+## Disabled-By-Default Rollout
+
+> **Rollout recommendation:** keep Hook Chains disabled until you validate rules in your environment.
+>
+> - Set top-level config to `"enabled": false` initially.
+> - Enable per environment when ready.
+> - Dispatch is gated by `feature('HOOK_CHAINS')`.
+> - Env gate defaults to off unless `CLAUDE_CODE_ENABLE_HOOK_CHAINS=1` is set.
+
+This keeps existing workflows unchanged while you tune guard windows and action behavior.
+
+## Feature Overview
+
+Hook Chains are loaded from a deterministic config file and evaluated on dispatched hook events.
+
+MVP runtime trigger wiring:
+
+- `PostToolUseFailure` hooks dispatch Hook Chains with outcome `failed`.
+- `TaskCompleted` hooks dispatch Hook Chains with outcome:
+  - `success` when completion hooks did not block.
+  - `failed` when completion hooks returned blocking errors or prevented continuation.
+
+Default config path:
+
+- `.openclaude/hook-chains.json`
+
+Override path:
+
+- `CLAUDE_CODE_HOOK_CHAINS_CONFIG_PATH=/abs/or/relative/path/to/hook-chains.json`
+
+Global gate:
+
+- `feature('HOOK_CHAINS')` must be enabled in the build
+- `CLAUDE_CODE_ENABLE_HOOK_CHAINS=0|1` (defaults to disabled when unset)
+
+## Safety Guarantees
+
+The runtime is intentionally conservative:
+
+- **Depth guard:** chain dispatch is blocked when `chainDepth >= maxChainDepth`.
+- **Rule cooldown:** each rule can only re-fire after cooldown expires.
+- **Dedup window:** identical event/action combinations are suppressed for a window.
+- **Abort-safe behavior:** if the current signal is aborted, actions skip safely.
+- **Policy-aware remote warm:** `warm_remote_capacity` skips when remote sessions are policy denied.
+- **Bridge inactive no-op:** `warm_remote_capacity` safely skips when no active bridge handle exists.
+- **Missing team context safety:** `notify_team` skips with structured reason if no team context/team file is available.
+- **Fallback launcher safety:** `spawn_fallback_agent` fails with a structured reason when launch permissions/context are unavailable.
+
+## Configuration Schema Reference
+
+Top-level object:
+
+```json
+{
+  "version": 1,
+  "enabled": true,
+  "maxChainDepth": 2,
+  "defaultCooldownMs": 30000,
+  "defaultDedupWindowMs": 30000,
+  "rules": []
+}
+```
+
+### Top-Level Fields
+
+| Field | Type | Required | Notes |
+|---|---|---:|---|
+| `version` | `1` | No | Defaults to `1`. |
+| `enabled` | `boolean` | No | Global feature switch for this config file. |
+| `maxChainDepth` | `integer` | No | Global depth guard (default `2`, max `10`). |
+| `defaultCooldownMs` | `integer` | No | Default rule cooldown in ms (default `30000`). |
+| `defaultDedupWindowMs` | `integer` | No | Default action dedup window in ms (default `30000`). |
+| `rules` | `HookChainRule[]` | No | Defaults to `[]`. May be omitted or empty; when no rules are present, dispatch is a no-op and returns `enabled: false`. |
+
+> **Note:** An empty ruleset is valid and can be used to keep Hook Chains configured but effectively disabled until rules are added.
+### Rule Object (`HookChainRule`)
+
+```json
+{
+  "id": "task-failure-recovery",
+  "enabled": true,
+  "trigger": {
+    "event": "TaskCompleted",
+    "outcome": "failed"
+  },
+  "condition": {
+    "toolNames": ["Edit"],
+    "taskStatuses": ["failed"],
+    "errorIncludes": ["timeout", "permission denied"],
+    "eventFieldEquals": {
+      "meta.source": "scheduler"
+    }
+  },
+  "cooldownMs": 60000,
+  "dedupWindowMs": 30000,
+  "maxDepth": 2,
+  "actions": []
+}
+```
+
+| Field | Type | Required | Notes |
+|---|---|---:|---|
+| `id` | `string` | Yes | Stable identifier used in telemetry/guards. |
+| `enabled` | `boolean` | No | Per-rule switch. |
+| `trigger.event` | `HookEvent` | Yes | Event name to match. |
+| `trigger.outcome` | `"success"|"failed"|"timeout"|"unknown"` | No | Single outcome matcher. |
+| `trigger.outcomes` | `Outcome[]` | No | Multi-outcome matcher. Use either `outcome` or `outcomes`. |
+| `condition` | `object` | No | Optional extra matching constraints. |
+| `cooldownMs` | `integer` | No | Overrides global cooldown for this rule. |
+| `dedupWindowMs` | `integer` | No | Overrides global dedup for this rule. |
+| `maxDepth` | `integer` | No | Per-rule depth cap. |
+| `actions` | `HookChainAction[]` | Yes | One or more actions to execute in order. |
+
+### Condition Fields
+
+| Field | Type | Notes |
+|---|---|---|
+| `toolNames` | `string[]` | Matches `tool_name` / `toolName` in event payload. |
+| `taskStatuses` | `string[]` | Matches `task_status` / `taskStatus` / `status`. |
+| `errorIncludes` | `string[]` | Case-insensitive substring match against `error` / `reason` / `message`. |
+| `eventFieldEquals` | `Record<string, string\|number\|boolean>` | Dot-path equality against payload (example: `"meta.source": "scheduler"`). |
+
+### Actions
+
+#### `spawn_fallback_agent`
+
+```json
+{
+  "type": "spawn_fallback_agent",
+  "id": "fallback-1",
+  "enabled": true,
+  "dedupWindowMs": 30000,
+  "description": "Fallback recovery for failed task",
+  "promptTemplate": "Recover task ${TASK_SUBJECT}. Event=${EVENT_NAME}, outcome=${OUTCOME}, error=${ERROR}. Payload=${PAYLOAD_JSON}",
+  "agentType": "general-purpose",
+  "model": "sonnet"
+}
+```
+
+#### `notify_team`
+
+```json
+{
+  "type": "notify_team",
+  "id": "notify-ops",
+  "enabled": true,
+  "dedupWindowMs": 30000,
+  "teamName": "mesh-team",
+  "recipients": ["*"],
+  "summary": "Hook chain ${RULE_ID} fired",
+  "messageTemplate": "Event=${EVENT_NAME} outcome=${OUTCOME}\nTask=${TASK_ID}\nError=${ERROR}\nPayload=${PAYLOAD_JSON}"
+}
+```
+
+#### `warm_remote_capacity`
+
+```json
+{
+  "type": "warm_remote_capacity",
+  "id": "warm-bridge",
+  "enabled": true,
+  "dedupWindowMs": 60000,
+  "createDefaultEnvironmentIfMissing": false
+}
+```
+
+## Complete Example Configs
+
+### 1) Retry via Fallback Agent
+
+```json
+{
+  "version": 1,
+  "enabled": true,
+  "maxChainDepth": 2,
+  "defaultCooldownMs": 30000,
+  "defaultDedupWindowMs": 30000,
+  "rules": [
+    {
+      "id": "retry-task-via-fallback",
+      "trigger": {
+        "event": "TaskCompleted",
+        "outcome": "failed"
+      },
+      "cooldownMs": 60000,
+      "actions": [
+        {
+          "type": "spawn_fallback_agent",
+          "id": "spawn-retry-agent",
+          "description": "Retry failed task with fallback agent",
+          "promptTemplate": "A task failed. Recover it safely.\nTask=${TASK_SUBJECT}\nDescription=${TASK_DESCRIPTION}\nError=${ERROR}\nPayload=${PAYLOAD_JSON}",
+          "agentType": "general-purpose",
+          "model": "sonnet"
+        }
+      ]
+    }
+  ]
+}
+```
+
+### 2) Notify Only
+
+```json
+{
+  "version": 1,
+  "enabled": true,
+  "maxChainDepth": 2,
+  "defaultCooldownMs": 30000,
+  "defaultDedupWindowMs": 30000,
+  "rules": [
+    {
+      "id": "notify-on-tool-failure",
+      "trigger": {
+        "event": "PostToolUseFailure",
+        "outcome": "failed"
+      },
+      "condition": {
+        "toolNames": ["Edit", "Write", "Bash"]
+      },
+      "actions": [
+        {
+          "type": "notify_team",
+          "id": "notify-team-failure",
+          "recipients": ["*"],
+          "summary": "Tool failure detected",
+          "messageTemplate": "Tool failure detected.\nEvent=${EVENT_NAME} outcome=${OUTCOME}\nError=${ERROR}\nPayload=${PAYLOAD_JSON}"
+        }
+      ]
+    }
+  ]
+}
+```
+
+### 3) Combined Fallback + Notify + Bridge Warm
+
+```json
+{
+  "version": 1,
+  "enabled": true,
+  "maxChainDepth": 2,
+  "defaultCooldownMs": 45000,
+  "defaultDedupWindowMs": 30000,
+  "rules": [
+    {
+      "id": "full-recovery-chain",
+      "trigger": {
+        "event": "TaskCompleted",
+        "outcomes": ["failed", "timeout"]
+      },
+      "condition": {
+        "errorIncludes": ["timeout", "capacity", "connection"]
+      },
+      "cooldownMs": 90000,
+      "actions": [
+        {
+          "type": "spawn_fallback_agent",
+          "id": "fallback-agent",
+          "description": "Recover failed task execution",
+          "promptTemplate": "Recover failed task and produce a concise fix summary.\nTask=${TASK_SUBJECT}\nError=${ERROR}\nPayload=${PAYLOAD_JSON}"
+        },
+        {
+          "type": "notify_team",
+          "id": "notify-team",
+          "recipients": ["*"],
+          "summary": "Recovery chain triggered",
+          "messageTemplate": "Recovery chain ${RULE_ID} fired.\nOutcome=${OUTCOME}\nTask=${TASK_SUBJECT}\nError=${ERROR}"
+        },
+        {
+          "type": "warm_remote_capacity",
+          "id": "warm-capacity",
+          "createDefaultEnvironmentIfMissing": false
+        }
+      ]
+    }
+  ]
+}
+```
+
+## Template Variables
+
+The following placeholders are supported by `promptTemplate`, `summary`, and `messageTemplate`:
+
+- `${EVENT_NAME}`
+- `${OUTCOME}`
+- `${RULE_ID}`
+- `${TASK_SUBJECT}`
+- `${TASK_DESCRIPTION}`
+- `${TASK_ID}`
+- `${ERROR}`
+- `${PAYLOAD_JSON}`
+
+## Troubleshooting
+
+### Rule never triggers
+
+- Verify `trigger.event` and `trigger.outcome`/`trigger.outcomes` exactly match dispatched event data.
+- Check `condition` filters (especially `toolNames` and `eventFieldEquals` dot-path keys).
+- Confirm the config file is valid JSON and schema-valid.
+
+### Actions show as skipped
+
+Common skip reasons:
+
+- `action disabled`
+- `rule cooldown active ...`
+- `dedup window active ...`
+- `max chain depth reached ...`
+- `No team context is available ...`
+- `Team file not found ...`
+- `Remote sessions are blocked by policy`
+- `Bridge is not active; warm_remote_capacity is a safe no-op`
+- `No fallback agent launcher is registered in runtime context`
+
+### Config changes not reflected
+
+- Loader uses memoization by file mtime/size.
+- Ensure your editor writes the file fully and updates mtime.
+- If needed, force reload from the caller side with `forceReloadConfig: true`.
+
+### Existing workflows changed unexpectedly
+
+- Set `"enabled": false` at top-level.
+- Or globally disable with `CLAUDE_CODE_ENABLE_HOOK_CHAINS=0`.
+- Re-enable gradually after validating one rule at a time.

docs/quick-start-mac-linux.md

@@ -41,11 +41,13 @@ openclaude
 export CLAUDE_CODE_USE_OPENAI=1
 export OPENAI_API_KEY=sk-your-key-here
 export OPENAI_BASE_URL=https://api.deepseek.com/v1
-export OPENAI_MODEL=deepseek-chat
+export OPENAI_MODEL=deepseek-v4-flash
 
 openclaude
 ```
 
+Use `deepseek-v4-pro` when you want the stronger model. `deepseek-chat` and `deepseek-reasoner` still work as DeepSeek's legacy API aliases.
+
 ### Option C: Ollama
 
 Install Ollama first from:

docs/quick-start-windows.md

@@ -41,11 +41,13 @@ openclaude
 $env:CLAUDE_CODE_USE_OPENAI="1"
 $env:OPENAI_API_KEY="sk-your-key-here"
 $env:OPENAI_BASE_URL="https://api.deepseek.com/v1"
-$env:OPENAI_MODEL="deepseek-chat"
+$env:OPENAI_MODEL="deepseek-v4-flash"
 
 openclaude
 ```
 
+Use `deepseek-v4-pro` when you want the stronger model. `deepseek-chat` and `deepseek-reasoner` still work as DeepSeek's legacy API aliases.
+
 ### Option C: Ollama
 
 Install Ollama first from:

package.json

@@ -1,7 +1,7 @@
 {
   "name": "@gitlawb/openclaude",
-  "version": "0.6.0",
-  "description": "Claude Code opened to any LLM — OpenAI, Gemini, DeepSeek, Ollama, and 200+ models",
+  "version": "0.7.0",
+  "description": "OpenClaude opens coding-agent workflows to any LLM — OpenAI, Gemini, DeepSeek, Ollama, and 200+ models",
   "type": "module",
   "bin": {
     "openclaude": "./bin/openclaude"

scripts/build.ts

@@ -34,6 +34,7 @@ const featureFlags: Record<string, boolean> = {
   WEB_BROWSER_TOOL: false,        // Built-in browser automation (source not mirrored)
   CHICAGO_MCP: false,             // Computer-use MCP (native Swift modules stubbed)
   COWORKER_TYPE_TELEMETRY: false, // Telemetry for agent/coworker type classification
+  MCP_SKILLS: false,              // Dynamic MCP skill discovery (src/skills/mcpSkills.ts not mirrored; enabling this causes "fetchMcpSkillsForClient is not a function" when MCP servers with resources connect — see #856)
 
   // ── Enabled: upstream defaults ──────────────────────────────────────
   COORDINATOR_MODE: true,             // Multi-agent coordinator with worker delegation
@@ -56,7 +57,6 @@ const featureFlags: Record<string, boolean> = {
   EXTRACT_MEMORIES: true,             // Auto-extract durable memories from conversations
   FORK_SUBAGENT: true,                // Implicit context-forking when omitting subagent_type
   VERIFICATION_AGENT: true,           // Built-in read-only agent for test/verification
-  MCP_SKILLS: true,                   // Discover skills dynamically from MCP server resources
   PROMPT_CACHE_BREAK_DETECTION: true, // Detect & log unexpected prompt cache invalidations
   HOOK_PROMPTS: true,                 // Allow tools to request interactive user prompts
 }

scripts/feature-flags-source-guard.test.ts

@@ -0,0 +1,47 @@
+import { existsSync, readFileSync } from 'fs'
+import { join } from 'path'
+import { expect, test } from 'bun:test'
+
+// Regression guard for #856. Several build feature flags require source files
+// that are not mirrored into the open build. When such a flag is set to `true`
+// without the source present, the bundler falls back to a missing-module stub
+// that only exports `default`, which causes runtime errors like
+// `fetchMcpSkillsForClient is not a function` when downstream code reaches
+// through the `require()` to a named export.
+//
+// This test fails fast at test-time if someone re-enables one of these flags
+// without first mirroring the corresponding source file.
+
+const BUILD_SCRIPT = join(import.meta.dir, 'build.ts')
+const REPO_ROOT = join(import.meta.dir, '..')
+
+type FlagGuard = {
+  flag: string
+  source: string // path relative to repo root
+}
+
+const FLAG_REQUIRES_SOURCE: FlagGuard[] = [
+  { flag: 'MCP_SKILLS', source: 'src/skills/mcpSkills.ts' },
+]
+
+test('build feature flags are not enabled without their source files', () => {
+  const buildScript = readFileSync(BUILD_SCRIPT, 'utf-8')
+
+  for (const { flag, source } of FLAG_REQUIRES_SOURCE) {
+    const enabledRe = new RegExp(`^\\s*${flag}\\s*:\\s*true\\b`, 'm')
+    const isEnabled = enabledRe.test(buildScript)
+    const sourceExists = existsSync(join(REPO_ROOT, source))
+
+    if (isEnabled && !sourceExists) {
+      throw new Error(
+        `Feature flag ${flag} is enabled in scripts/build.ts, but its required source file "${source}" does not exist. ` +
+          `Enabling this flag without the source will cause runtime errors (missing named exports from the missing-module stub). ` +
+          `Either mirror the source file or set ${flag}: false.`,
+      )
+    }
+
+    // When the source IS present, the flag can be either true or false; either
+    // is fine. We only care about the "enabled but missing" combination.
+    expect(true).toBe(true)
+  }
+})

src/Tool.ts

@@ -249,6 +249,11 @@ export type ToolUseContext = {
   /** When true, canUseTool must always be called even when hooks auto-approve.
    *  Used by speculation for overlay file path rewriting. */
   requireCanUseTool?: boolean
+  /**
+   * Optional callback used by hook-chain fallback actions that launch
+   * AgentTool from hook runtime paths.
+   */
+  hookChainsCanUseTool?: CanUseToolFn
   messages: Message[]
   fileReadingLimits?: {
     maxTokens?: number

src/__tests__/bugfixes.test.ts

@@ -169,6 +169,14 @@ describe('Web search result count improvements', () => {
 
     expect(content).toMatch(/max_uses:\s*15/)
   })
+
+  test('codex web search path guarantees a non-empty result body', async () => {
+    const content = await file(
+      'tools/WebSearchTool/WebSearchTool.ts',
+    ).text()
+
+    expect(content).toContain("results.push('No results found.')")
+  })
 })
 
 // ---------------------------------------------------------------------------

src/bridge/bridgeEnabled.ts

@@ -70,13 +70,13 @@ export async function isBridgeEnabledBlocking(): Promise<boolean> {
 export async function getBridgeDisabledReason(): Promise<string | null> {
   if (feature('BRIDGE_MODE')) {
     if (!isClaudeAISubscriber()) {
-      return 'Remote Control requires a claude.ai subscription. Run `claude auth login` to sign in with your claude.ai account.'
+      return 'Remote Control requires a claude.ai subscription. Run `openclaude auth login` to sign in with your claude.ai account.'
     }
     if (!hasProfileScope()) {
-      return 'Remote Control requires a full-scope login token. Long-lived tokens (from `claude setup-token` or CLAUDE_CODE_OAUTH_TOKEN) are limited to inference-only for security reasons. Run `claude auth login` to use Remote Control.'
+      return 'Remote Control requires a full-scope login token. Long-lived tokens (from `openclaude setup-token` or CLAUDE_CODE_OAUTH_TOKEN) are limited to inference-only for security reasons. Run `openclaude auth login` to use Remote Control.'
     }
     if (!getOauthAccountInfo()?.organizationUuid) {
-      return 'Unable to determine your organization for Remote Control eligibility. Run `claude auth login` to refresh your account information.'
+      return 'Unable to determine your organization for Remote Control eligibility. Run `openclaude auth login` to refresh your account information.'
     }
     if (!(await checkGate_CACHED_OR_BLOCKING('tengu_ccr_bridge'))) {
       return 'Remote Control is not yet enabled for your account.'
@@ -166,7 +166,7 @@ export function checkBridgeMinVersion(): string | null {
       minVersion: string
     }>('tengu_bridge_min_version', { minVersion: '0.0.0' })
     if (config.minVersion && lt(MACRO.VERSION, config.minVersion)) {
-      return `Your version of Claude Code (${MACRO.VERSION}) is too old for Remote Control.\nVersion ${config.minVersion} or higher is required. Run \`claude update\` to update.`
+      return `Your version of OpenClaude (${MACRO.VERSION}) is too old for Remote Control.\nVersion ${config.minVersion} or higher is required. Run \`openclaude update\` to update.`
     }
   }
   return null

src/bridge/bridgeMain.ts

@@ -2248,7 +2248,7 @@ export async function bridgeMain(args: string[]): Promise<void> {
     })
     // biome-ignore lint/suspicious/noConsole: intentional dialog output
     console.log(
-      `\nClaude Remote Control is launching in spawn mode which lets you create new sessions in this project from Claude Code on Web or your Mobile app. Learn more here: https://code.claude.com/docs/en/remote-control\n\n` +
+      `\nClaude Remote Control is launching in spawn mode which lets you create new sessions in this project from OpenClaude on the web or your mobile app. Learn more here: https://code.claude.com/docs/en/remote-control\n\n` +
         `Spawn mode for this project:\n` +
         `  [1] same-dir \u2014 sessions share the current directory (default)\n` +
         `  [2] worktree \u2014 each session gets an isolated git worktree\n\n` +

src/bridge/envLessBridgeConfig.ts

@@ -147,7 +147,7 @@ export async function getEnvLessBridgeConfig(): Promise<EnvLessBridgeConfig> {
 export async function checkEnvLessBridgeMinVersion(): Promise<string | null> {
   const cfg = await getEnvLessBridgeConfig()
   if (cfg.min_version && lt(MACRO.VERSION, cfg.min_version)) {
-    return `Your version of Claude Code (${MACRO.VERSION}) is too old for Remote Control.\nVersion ${cfg.min_version} or higher is required. Run \`claude update\` to update.`
+    return `Your version of OpenClaude (${MACRO.VERSION}) is too old for Remote Control.\nVersion ${cfg.min_version} or higher is required. Run \`openclaude update\` to update.`
   }
   return null
 }

src/bridge/initReplBridge.ts

@@ -415,7 +415,7 @@ export async function initReplBridge(
         `[bridge:repl] Skipping: ${versionError}`,
         true,
       )
-      onStateChange?.('failed', 'run `claude update` to upgrade')
+      onStateChange?.('failed', 'run `openclaude update` to upgrade')
       return null
     }
     logForDebugging(
@@ -456,7 +456,7 @@ export async function initReplBridge(
   const versionError = checkBridgeMinVersion()
   if (versionError) {
     logBridgeSkip('version_too_old', `[bridge:repl] Skipping: ${versionError}`)
-    onStateChange?.('failed', 'run `claude update` to upgrade')
+    onStateChange?.('failed', 'run `openclaude update` to upgrade')
     return null
   }
 

src/bridge/trustedDevice.ts

@@ -147,7 +147,7 @@ export async function enrollTrustedDevice(): Promise<void> {
         device_id?: string
       }>(
         `${baseUrl}/api/auth/trusted_devices`,
-        { display_name: `Claude Code on ${hostname()} · ${process.platform}` },
+        { display_name: `OpenClaude on ${hostname()} · ${process.platform}` },
         {
           headers: {
             Authorization: `Bearer ${accessToken}`,

src/cli/handlers/auth.ts

@@ -287,7 +287,7 @@ export async function authStatus(opts: {
     }
     if (!loggedIn) {
       process.stdout.write(
-        'Not logged in. Run claude auth login to authenticate.\n',
+        'Not logged in. Run openclaude auth login to authenticate.\n',
       )
     }
   } else {

src/cli/handlers/autoMode.ts

@@ -83,7 +83,7 @@ export async function autoModeCritiqueHandler(options: {
     process.stdout.write(
       'No custom auto mode rules found.\n\n' +
         'Add rules to your settings file under autoMode.{allow, soft_deny, environment}.\n' +
-        'Run `claude auto-mode defaults` to see the default rules for reference.\n',
+        'Run `openclaude auto-mode defaults` to see the default rules for reference.\n',
     )
     return
   }

src/cli/handlers/mcp.tsx

@@ -233,7 +233,7 @@ export async function mcpRemoveHandler(name: string, options: {
       });
       process.stderr.write('\nTo remove from a specific scope, use:\n');
       scopes.forEach(scope => {
-        process.stderr.write(`  claude mcp remove "${name}" -s ${scope}\n`);
+        process.stderr.write(`  openclaude mcp remove "${name}" -s ${scope}\n`);
       });
       cliError();
     }
@@ -250,7 +250,7 @@ export async function mcpListHandler(): Promise<void> {
   } = await getAllMcpConfigs();
   if (Object.keys(configs).length === 0) {
     // biome-ignore lint/suspicious/noConsole:: intentional console output
-    console.log('No MCP servers configured. Use `claude mcp add` to add a server.');
+    console.log('No MCP servers configured. Use `openclaude mcp add` to add a server.');
   } else {
     // biome-ignore lint/suspicious/noConsole:: intentional console output
     console.log('Checking MCP server health...\n');
@@ -374,7 +374,7 @@ export async function mcpGetHandler(name: string): Promise<void> {
     }
   }
   // biome-ignore lint/suspicious/noConsole:: intentional console output
-  console.log(`\nTo remove this server, run: claude mcp remove "${name}" -s ${server.scope}`);
+  console.log(`\nTo remove this server, run: openclaude mcp remove "${name}" -s ${server.scope}`);
   // Use gracefulShutdown to properly clean up MCP server connections
   // (process.exit bypasses cleanup handlers, leaving child processes orphaned)
   await gracefulShutdown(0);
@@ -455,5 +455,5 @@ export async function mcpResetChoicesHandler(): Promise<void> {
     disabledMcpjsonServers: [],
     enableAllProjectMcpServers: false
   }));
-  cliOk('All project-scoped (.mcp.json) server approvals and rejections have been reset.\n' + 'You will be prompted for approval next time you start Claude Code.');
+  cliOk('All project-scoped (.mcp.json) server approvals and rejections have been reset.\n' + 'You will be prompted for approval next time you start OpenClaude.');
 }

src/cli/handlers/plugins.ts

@@ -352,7 +352,7 @@ export async function pluginListHandler(options: {
     // through to the session section so the failure is visible.
     if (inlineLoadErrors.length === 0) {
       cliOk(
-        'No plugins installed. Use `claude plugin install` to install a plugin.',
+        'No plugins installed. Use `openclaude plugin install` to install a plugin.',
       )
     }
   }

src/cli/print.ts

@@ -5026,7 +5026,7 @@ async function loadInitialMessages(
       )
       if (!parsedSessionId) {
         let errorMessage =
-          'Error: --resume requires a valid session ID when used with --print. Usage: claude -p --resume <session-id>'
+          'Error: --resume requires a valid session ID when used with --print. Usage: openclaude -p --resume <session-id>'
         if (typeof options.resume === 'string') {
           errorMessage += `. Session IDs must be in UUID format (e.g., 550e8400-e29b-41d4-a716-446655440000). Provided value "${options.resume}" is not a valid UUID`
         }

src/cli/update.ts

@@ -35,15 +35,20 @@ export async function update() {
   // binary (without it).
   if (getAPIProvider() !== 'firstParty') {
     writeToStdout(
-      chalk.yellow('Auto-update is not available for third-party provider builds.\n') +
-      'To update, pull the latest source from the repository and rebuild:\n' +
-      '  git pull && bun install && bun run build\n',
+      chalk.yellow(
+        `Auto-update is not available for third-party provider builds.\n`,
+      ) +
+        `Current version: ${MACRO.DISPLAY_VERSION}\n\n` +
+        `To update, reinstall from npm:\n` +
+        chalk.bold(`  npm install -g ${MACRO.PACKAGE_URL}@latest`) + '\n\n' +
+        `Or, if you built from source, pull and rebuild:\n` +
+        chalk.bold('  git pull && bun install && bun run build') + '\n',
     )
-    return
+    await gracefulShutdown(0)
   }
 
   logEvent('tengu_update_check', {})
-  writeToStdout(`Current version: ${MACRO.VERSION}\n`)
+  writeToStdout(`Current version: ${MACRO.DISPLAY_VERSION}\n`)
 
   const channel = getInitialSettings()?.autoUpdatesChannel ?? 'latest'
   writeToStdout(`Checking for updates to ${channel} version...\n`)
@@ -123,9 +128,14 @@ export async function update() {
   if (diagnostic.installationType === 'development') {
     writeToStdout('\n')
     writeToStdout(
-      chalk.yellow('Warning: Cannot update development build') + '\n',
+      chalk.yellow('You are running a development build — auto-update is unavailable.') + '\n',
     )
-    await gracefulShutdown(1)
+    writeToStdout('To update, pull the latest source and rebuild:\n')
+    writeToStdout(chalk.bold('  git pull && bun install && bun run build') + '\n')
+    writeToStdout('\n')
+    writeToStdout('Or reinstall from npm:\n')
+    writeToStdout(chalk.bold(`  npm install -g ${MACRO.PACKAGE_URL}@latest`) + '\n')
+    await gracefulShutdown(0)
   }
 
   // Check if running from a package manager
@@ -136,8 +146,8 @@ export async function update() {
     if (packageManager === 'homebrew') {
       writeToStdout('Claude is managed by Homebrew.\n')
       const latest = await getLatestVersion(channel)
-      if (latest && !gte(MACRO.VERSION, latest)) {
-        writeToStdout(`Update available: ${MACRO.VERSION} → ${latest}\n`)
+      if (latest && !gte(MACRO.DISPLAY_VERSION, latest)) {
+        writeToStdout(`Update available: ${MACRO.DISPLAY_VERSION} → ${latest}\n`)
         writeToStdout('\n')
         writeToStdout('To update, run:\n')
         writeToStdout(chalk.bold('  brew upgrade claude-code') + '\n')
@@ -147,8 +157,8 @@ export async function update() {
     } else if (packageManager === 'winget') {
       writeToStdout('Claude is managed by winget.\n')
       const latest = await getLatestVersion(channel)
-      if (latest && !gte(MACRO.VERSION, latest)) {
-        writeToStdout(`Update available: ${MACRO.VERSION} → ${latest}\n`)
+      if (latest && !gte(MACRO.DISPLAY_VERSION, latest)) {
+        writeToStdout(`Update available: ${MACRO.DISPLAY_VERSION} → ${latest}\n`)
         writeToStdout('\n')
         writeToStdout('To update, run:\n')
         writeToStdout(
@@ -160,8 +170,8 @@ export async function update() {
     } else if (packageManager === 'apk') {
       writeToStdout('Claude is managed by apk.\n')
       const latest = await getLatestVersion(channel)
-      if (latest && !gte(MACRO.VERSION, latest)) {
-        writeToStdout(`Update available: ${MACRO.VERSION} → ${latest}\n`)
+      if (latest && !gte(MACRO.DISPLAY_VERSION, latest)) {
+        writeToStdout(`Update available: ${MACRO.DISPLAY_VERSION} → ${latest}\n`)
         writeToStdout('\n')
         writeToStdout('To update, run:\n')
         writeToStdout(chalk.bold('  apk upgrade claude-code') + '\n')
@@ -250,14 +260,14 @@ export async function update() {
         await gracefulShutdown(1)
       }
 
-      if (result.latestVersion === MACRO.VERSION) {
+      if (result.latestVersion === MACRO.DISPLAY_VERSION) {
         writeToStdout(
-          chalk.green(`Claude Code is up to date (${MACRO.VERSION})`) + '\n',
+          chalk.green(`OpenClaude is up to date (${MACRO.DISPLAY_VERSION})`) + '\n',
         )
       } else {
         writeToStdout(
           chalk.green(
-            `Successfully updated from ${MACRO.VERSION} to version ${result.latestVersion}`,
+            `Successfully updated from ${MACRO.DISPLAY_VERSION} to version ${result.latestVersion}`,
           ) + '\n',
         )
         await regenerateCompletionCache()
@@ -266,7 +276,7 @@ export async function update() {
     } catch (error) {
       process.stderr.write('Error: Failed to install native update\n')
       process.stderr.write(String(error) + '\n')
-      process.stderr.write('Try running "claude doctor" for diagnostics\n')
+      process.stderr.write('Try running "openclaude doctor" for diagnostics\n')
       await gracefulShutdown(1)
     }
   }
@@ -320,15 +330,15 @@ export async function update() {
   }
 
   // Check if versions match exactly, including any build metadata (like SHA)
-  if (latestVersion === MACRO.VERSION) {
+  if (latestVersion === MACRO.DISPLAY_VERSION) {
     writeToStdout(
-      chalk.green(`Claude Code is up to date (${MACRO.VERSION})`) + '\n',
+      chalk.green(`OpenClaude is up to date (${MACRO.DISPLAY_VERSION})`) + '\n',
     )
     await gracefulShutdown(0)
   }
 
   writeToStdout(
-    `New version available: ${latestVersion} (current: ${MACRO.VERSION})\n`,
+    `New version available: ${latestVersion} (current: ${MACRO.DISPLAY_VERSION})\n`,
   )
   writeToStdout('Installing update...\n')
 
@@ -388,7 +398,7 @@ export async function update() {
     case 'success':
       writeToStdout(
         chalk.green(
-          `Successfully updated from ${MACRO.VERSION} to version ${latestVersion}`,
+          `Successfully updated from ${MACRO.DISPLAY_VERSION} to version ${latestVersion}`,
         ) + '\n',
       )
       await regenerateCompletionCache()

src/commands.ts

@@ -21,6 +21,7 @@ import dream from './commands/dream/index.js'
 import ctx_viz from './commands/ctx_viz/index.js'
 import doctor from './commands/doctor/index.js'
 import onboardGithub from './commands/onboard-github/index.js'
+import knowledge from './commands/knowledge/index.js'
 import memory from './commands/memory/index.js'
 import help from './commands/help/index.js'
 import ide from './commands/ide/index.js'
@@ -33,6 +34,7 @@ import installGitHubApp from './commands/install-github-app/index.js'
 import installSlackApp from './commands/install-slack-app/index.js'
 import breakCache from './commands/break-cache/index.js'
 import cacheProbe from './commands/cache-probe/index.js'
+import cacheStats from './commands/cacheStats/index.js'
 import mcp from './commands/mcp/index.js'
 import mobile from './commands/mobile/index.js'
 import onboarding from './commands/onboarding/index.js'
@@ -197,7 +199,7 @@ import stats from './commands/stats/index.js'
 const usageReport: Command = {
   type: 'prompt',
   name: 'insights',
-  description: 'Generate a report analyzing your Claude Code sessions',
+  description: 'Generate a report analyzing your OpenClaude sessions',
   contentLength: 0,
   progressMessage: 'analyzing your sessions',
   source: 'builtin',
@@ -270,6 +272,7 @@ const COMMANDS = memoize((): Command[] => [
   branch,
   btw,
   cacheProbe,
+  cacheStats,
   chrome,
   clear,
   color,
@@ -292,6 +295,7 @@ const COMMANDS = memoize((): Command[] => [
   ide,
   init,
   keybindings,
+  knowledge,
   installGitHubApp,
   installSlackApp,
   mcp,

src/commands/buddy/index.ts

@@ -3,7 +3,7 @@ import type { Command } from '../../commands.js'
 const buddy = {
   type: 'local-jsx',
   name: 'buddy',
-  description: 'Hatch, pet, and manage your Open Claude companion',
+  description: 'Hatch, pet, and manage your OpenClaude companion',
   immediate: true,
   argumentHint: '[status|mute|unmute|help]',
   load: () => import('./buddy.js'),

src/commands/cacheStats/cacheStats.test.ts

@@ -0,0 +1,157 @@
+/**
+ * Tests for `/cache-stats` command rendering.
+ *
+ * The command has non-trivial string formatting (timestamp slicing, model
+ * label padding, conditional N/A footnote, recent-rows cap) which can
+ * silently regress — these snapshot tests keep it honest.
+ */
+import { beforeEach, describe, expect, test } from 'bun:test'
+import type { CacheMetrics } from '../../services/api/cacheMetrics.js'
+import {
+  _setHistoryCapForTesting,
+  recordRequest,
+  resetSessionCacheStats,
+} from '../../services/api/cacheStatsTracker.js'
+import { call } from './cacheStats.js'
+
+function supported(partial: Partial<CacheMetrics>): CacheMetrics {
+  return {
+    read: 0,
+    created: 0,
+    total: 0,
+    hitRate: null,
+    supported: true,
+    ...partial,
+  }
+}
+
+const UNSUPPORTED: CacheMetrics = {
+  read: 0,
+  created: 0,
+  total: 0,
+  hitRate: null,
+  supported: false,
+}
+
+// The command signature requires a LocalJSXCommandContext. Our command
+// doesn't actually read it — we pass an empty stand-in so the test can
+// invoke call() without dragging the whole REPL context in.
+const EMPTY_CTX = {} as Parameters<typeof call>[1]
+
+// /cache-stats always returns a text result. Narrow the union here so
+// the assertions don't need to redo the discriminant check every call.
+async function runCommand(): Promise<string> {
+  const result = await call('', EMPTY_CTX)
+  if (result.type !== 'text') {
+    throw new Error(
+      `cacheStats command must return type:'text', got ${result.type}`,
+    )
+  }
+  return result.value
+}
+
+beforeEach(() => {
+  resetSessionCacheStats()
+  _setHistoryCapForTesting(500)
+})
+
+describe('/cache-stats — empty session', () => {
+  test('shows friendly "no requests yet" message', async () => {
+    const value = await runCommand()
+    expect(value).toContain('No API requests yet this session')
+    expect(value).toContain('/cache-stats')
+  })
+})
+
+describe('/cache-stats — supported-only session', () => {
+  test('renders Cache stats header, turn and session summaries', async () => {
+    recordRequest(
+      supported({ read: 500, total: 1_000, hitRate: 0.5 }),
+      'claude-sonnet-4',
+    )
+    const value = await runCommand()
+    expect(value).toContain('Cache stats')
+    expect(value).toContain('Current turn:')
+    expect(value).toContain('Session total:')
+    // Compact metric line should appear in the recent-requests table.
+    expect(value).toContain('claude-sonnet-4')
+    expect(value).toContain('read')
+  })
+
+  test('omits the N/A footnote when every row is supported', async () => {
+    recordRequest(supported({ read: 200, total: 400, hitRate: 0.5 }), 'model-A')
+    const value = await runCommand()
+    expect(value).not.toContain('N/A rows')
+  })
+})
+
+describe('/cache-stats — mixed supported + unsupported', () => {
+  test('renders N/A footnote when any row is unsupported', async () => {
+    recordRequest(UNSUPPORTED, 'gpt-4-copilot')
+    recordRequest(
+      supported({ read: 100, total: 500, hitRate: 0.2 }),
+      'claude-sonnet-4',
+    )
+    const value = await runCommand()
+    expect(value).toContain(
+      'N/A rows: provider API does not expose cache usage',
+    )
+    expect(value).toContain('GitHub Copilot')
+    expect(value).toContain('Ollama')
+  })
+})
+
+describe('/cache-stats — recent-rows cap', () => {
+  test('caps the breakdown at 20 rows and reports omitted count', async () => {
+    for (let i = 0; i < 25; i++) {
+      recordRequest(
+        supported({ read: i, total: 100, hitRate: i / 100 }),
+        `model-${i}`,
+      )
+    }
+    const value = await runCommand()
+    // 20 shown, 5 omitted from the oldest end.
+    expect(value).toContain('(20 of 25, 5 older omitted)')
+    // Oldest rows (model-0..model-4) should not appear; newest must.
+    expect(value).toContain('model-24')
+    expect(value).not.toContain('model-0 ')
+  })
+
+  test('does not mention "older omitted" when all rows fit', async () => {
+    for (let i = 0; i < 5; i++) {
+      recordRequest(supported({ read: i, total: 10 }), `m${i}`)
+    }
+    const value = await runCommand()
+    expect(value).not.toContain('older omitted')
+    expect(value).toContain('(5)')
+  })
+})
+
+describe('/cache-stats — model label rendering', () => {
+  test('truncates long model labels to fit the column width', async () => {
+    // cacheStats.ts pads+slices the label to 28 chars for alignment.
+    const longLabel = 'some-extremely-long-model-identifier-that-wraps'
+    recordRequest(supported({ read: 10, total: 100, hitRate: 0.1 }), longLabel)
+    const value = await runCommand()
+    // Sliced to 28 chars.
+    expect(value).toContain(longLabel.slice(0, 28))
+    // And the full string should NOT appear (would mean no truncation).
+    expect(value).not.toContain(longLabel)
+  })
+})
+
+describe('/cache-stats — timestamp rendering', () => {
+  test('renders each row with full date and time (YYYY-MM-DD HH:MM:SS)', async () => {
+    recordRequest(supported({ read: 5, total: 10, hitRate: 0.5 }), 'claude-x')
+    const value = await runCommand()
+    // Match the full ISO-ish date + time the row uses. We assert the shape,
+    // not a specific timestamp — real clock is used, so a regex on the
+    // format is the right assertion.
+    expect(value).toMatch(/\d{4}-\d{2}-\d{2} \d{2}:\d{2}:\d{2}/)
+    // Bare time-of-day alone (no date) should NOT appear in isolation — it
+    // must always be preceded by the date. Guards against regression if
+    // someone shortens the formatter again.
+    const timeOnlyInRow = /\n\s*#\s*\d+\s+\d{2}:\d{2}:\d{2}\s/.test(value)
+    expect(timeOnlyInRow).toBe(false)
+  })
+})

src/commands/cacheStats/cacheStats.ts

@@ -0,0 +1,74 @@
+import {
+  getCacheStatsHistory,
+  getCurrentTurnCacheMetrics,
+  getSessionCacheMetrics,
+  type CacheStatsEntry,
+} from '../../services/api/cacheStatsTracker.js'
+import {
+  formatCacheMetricsCompact,
+  formatCacheMetricsFull,
+  type CacheMetrics,
+} from '../../services/api/cacheMetrics.js'
+import type { LocalCommandCall } from '../../types/command.js'
+
+// Cap the per-request breakdown to keep output readable. Users wanting
+// the full history can rely on OPENCLAUDE_LOG_TOKEN_USAGE=verbose for
+// structured per-request stderr output.
+const MAX_RECENT_ROWS = 20
+
+function formatRow(entry: CacheStatsEntry, idx: number): string {
+  // `YYYY-MM-DD HH:MM:SS` — long-running sessions can span midnight and a
+  // bare time-of-day makes the wrong row look "most recent" when two
+  // entries on different days share the same HH:MM:SS.
+  const iso = new Date(entry.timestamp).toISOString()
+  const ts = `${iso.slice(0, 10)} ${iso.slice(11, 19)}`
+  const line = formatCacheMetricsCompact(entry.metrics)
+  return `  #${String(idx + 1).padStart(3)}  ${ts}  ${entry.label.padEnd(28).slice(0, 28)}  ${line}`
+}
+
+function summarize(label: string, m: CacheMetrics): string {
+  return `${label.padEnd(18)}${formatCacheMetricsFull(m)}`
+}
+
+export const call: LocalCommandCall = async () => {
+  const history = getCacheStatsHistory()
+  const session = getSessionCacheMetrics()
+  const turn = getCurrentTurnCacheMetrics()
+
+  if (history.length === 0) {
+    return {
+      type: 'text',
+      value:
+        'Cache stats\n  No API requests yet this session.\n  Start a turn and re-run /cache-stats to see results.',
+    }
+  }
+
+  const recent = history.slice(-MAX_RECENT_ROWS)
+  const omitted = history.length - recent.length
+
+  const lines: string[] = ['Cache stats', '']
+  lines.push(summarize('Current turn:', turn))
+  lines.push(summarize('Session total:', session))
+  lines.push('')
+  lines.push(`Recent requests (${recent.length}${omitted > 0 ? ` of ${history.length}, ${omitted} older omitted` : ''}):`)
+  lines.push(`  #     time      model                         cache`)
+  for (const [i, entry] of recent.entries()) {
+    lines.push(formatRow(entry, history.length - recent.length + i))
+  }
+
+  // Honesty footnote — providers without cache reporting (vanilla Copilot,
+  // Ollama) show [Cache: N/A] rather than a fake 0%. Tell the user so they
+  // don't read "N/A" as "broken".
+  const hasUnsupported = recent.some((e) => !e.metrics.supported)
+  if (hasUnsupported) {
+    lines.push('')
+    lines.push(
+      '  N/A rows: provider API does not expose cache usage (GitHub Copilot, Ollama).',
+    )
+    lines.push(
+      '  The request still ran normally — only the metric is unavailable.',
+    )
+  }
+
+  return { type: 'text', value: lines.join('\n') }
+}

src/commands/cacheStats/index.ts

@@ -0,0 +1,24 @@
+/**
+ * /cache-stats — per-session cache diagnostics.
+ *
+ * Always-on diagnostic command (no toggle) that surfaces the metrics
+ * tracked in `cacheStatsTracker.ts`. Breaks cache usage down by request
+ * and also reports the session-wide aggregate — useful when the user
+ * suspects a cache bust (e.g. after /reload-plugins) and wants to see
+ * whether recent turns still hit the cache.
+ *
+ * Lazy-loaded (implementation in cacheStats.ts) to keep startup time
+ * minimal — same pattern used by /cost and /cache-probe.
+ */
+import type { Command } from '../../commands.js'
+
+const cacheStats = {
+  type: 'local',
+  name: 'cache-stats',
+  description:
+    'Show per-turn and session cache hit/miss stats (works across all providers)',
+  supportsNonInteractive: true,
+  load: () => import('./cacheStats.js'),
+} satisfies Command
+
+export default cacheStats

src/commands/chrome/chrome.tsx

@@ -197,7 +197,7 @@ function ClaudeInChromeMenu(t0) {
   }
   let t6;
   if ($[20] === Symbol.for("react.memo_cache_sentinel")) {
-    t6 = <Text>Claude in Chrome works with the Chrome extension to let you control your browser directly from Claude Code. Navigate websites, fill forms, capture screenshots, record GIFs, and debug with console logs and network requests.</Text>;
+    t6 = <Text>Claude in Chrome works with the Chrome extension to let you control your browser directly from OpenClaude. Navigate websites, fill forms, capture screenshots, record GIFs, and debug with console logs and network requests.</Text>;
     $[20] = t6;
   } else {
     t6 = $[20];

src/commands/createMovedToPluginCommand.ts

@@ -48,7 +48,7 @@ export function createMovedToPluginCommand({
             text: `This command has been moved to a plugin. Tell the user:
 
 1. To install the plugin, run:
-   claude plugin install ${pluginName}@claude-code-marketplace
+   openclaude plugin install ${pluginName}@claude-code-marketplace
 
 2. After installation, use /${pluginName}:${pluginCommand} to run this command
 

src/commands/doctor/index.ts

@@ -3,7 +3,7 @@ import { isEnvTruthy } from '../../utils/envUtils.js'
 
 const doctor: Command = {
   name: 'doctor',
-  description: 'Diagnose and verify your Claude Code installation and settings',
+  description: 'Diagnose and verify your OpenClaude installation and settings',
   isEnabled: () => !isEnvTruthy(process.env.DISABLE_DOCTOR_COMMAND),
   type: 'local-jsx',
   load: () => import('./doctor.js'),

src/commands/feedback/index.ts

@@ -7,7 +7,7 @@ const feedback = {
   aliases: ['bug'],
   type: 'local-jsx',
   name: 'feedback',
-  description: `Submit feedback about Claude Code`,
+  description: `Submit feedback about OpenClaude`,
   argumentHint: '[report]',
   isEnabled: () =>
     !(

src/commands/insights.ts

@@ -247,7 +247,7 @@ function getSessionMetaDir(): string {
   return join(getDataDir(), 'session-meta')
 }
 
-const FACET_EXTRACTION_PROMPT = `Analyze this Claude Code session and extract structured facets.
+const FACET_EXTRACTION_PROMPT = `Analyze this OpenClaude session and extract structured facets.
 
 CRITICAL GUIDELINES:
 
@@ -687,7 +687,7 @@ function formatTranscriptForFacets(log: LogOption): string {
   return lines.join('\n')
 }
 
-const SUMMARIZE_CHUNK_PROMPT = `Summarize this portion of a Claude Code session transcript. Focus on:
+const SUMMARIZE_CHUNK_PROMPT = `Summarize this portion of a OpenClaude session transcript. Focus on:
 1. What the user asked for
 2. What Claude did (tools used, files modified)
 3. Any friction or issues
@@ -1156,12 +1156,12 @@ type InsightSection = {
 const INSIGHT_SECTIONS: InsightSection[] = [
   {
     name: 'project_areas',
-    prompt: `Analyze this Claude Code usage data and identify project areas.
+    prompt: `Analyze this OpenClaude usage data and identify project areas.
 
 RESPOND WITH ONLY A VALID JSON OBJECT:
 {
   "areas": [
-    {"name": "Area name", "session_count": N, "description": "2-3 sentences about what was worked on and how Claude Code was used."}
+    {"name": "Area name", "session_count": N, "description": "2-3 sentences about what was worked on and how OpenClaude was used."}
   ]
 }
 
@@ -1170,18 +1170,18 @@ Include 4-5 areas. Skip internal CC operations.`,
   },
   {
     name: 'interaction_style',
-    prompt: `Analyze this Claude Code usage data and describe the user's interaction style.
+    prompt: `Analyze this OpenClaude usage data and describe the user's interaction style.
 
 RESPOND WITH ONLY A VALID JSON OBJECT:
 {
-  "narrative": "2-3 paragraphs analyzing HOW the user interacts with Claude Code. Use second person 'you'. Describe patterns: iterate quickly vs detailed upfront specs? Interrupt often or let Claude run? Include specific examples. Use **bold** for key insights.",
+  "narrative": "2-3 paragraphs analyzing HOW the user interacts with OpenClaude. Use second person 'you'. Describe patterns: iterate quickly vs detailed upfront specs? Interrupt often or let Claude run? Include specific examples. Use **bold** for key insights.",
   "key_pattern": "One sentence summary of most distinctive interaction style"
 }`,
     maxTokens: 8192,
   },
   {
     name: 'what_works',
-    prompt: `Analyze this Claude Code usage data and identify what's working well for this user. Use second person ("you").
+    prompt: `Analyze this OpenClaude usage data and identify what's working well for this user. Use second person ("you").
 
 RESPOND WITH ONLY A VALID JSON OBJECT:
 {
@@ -1196,7 +1196,7 @@ Include 3 impressive workflows.`,
   },
   {
     name: 'friction_analysis',
-    prompt: `Analyze this Claude Code usage data and identify friction points for this user. Use second person ("you").
+    prompt: `Analyze this OpenClaude usage data and identify friction points for this user. Use second person ("you").
 
 RESPOND WITH ONLY A VALID JSON OBJECT:
 {
@@ -1211,7 +1211,7 @@ Include 3 friction categories with 2 examples each.`,
   },
   {
     name: 'suggestions',
-    prompt: `Analyze this Claude Code usage data and suggest improvements.
+    prompt: `Analyze this OpenClaude usage data and suggest improvements.
 
 ## CC FEATURES REFERENCE (pick from these for features_to_try):
 1. **MCP Servers**: Connect Claude to external tools, databases, and APIs via Model Context Protocol.
@@ -1254,7 +1254,7 @@ IMPORTANT for features_to_try: Pick 2-3 from the CC FEATURES REFERENCE above. In
   },
   {
     name: 'on_the_horizon',
-    prompt: `Analyze this Claude Code usage data and identify future opportunities.
+    prompt: `Analyze this OpenClaude usage data and identify future opportunities.
 
 RESPOND WITH ONLY A VALID JSON OBJECT:
 {
@@ -1271,7 +1271,7 @@ Include 3 opportunities. Think BIG - autonomous workflows, parallel agents, iter
     ? [
         {
           name: 'cc_team_improvements',
-          prompt: `Analyze this Claude Code usage data and suggest product improvements for the CC team.
+          prompt: `Analyze this OpenClaude usage data and suggest product improvements for the CC team.
 
 RESPOND WITH ONLY A VALID JSON OBJECT:
 {
@@ -1285,7 +1285,7 @@ Include 2-3 improvements based on friction patterns observed.`,
         },
         {
           name: 'model_behavior_improvements',
-          prompt: `Analyze this Claude Code usage data and suggest model behavior improvements.
+          prompt: `Analyze this OpenClaude usage data and suggest model behavior improvements.
 
 RESPOND WITH ONLY A VALID JSON OBJECT:
 {
@@ -1301,7 +1301,7 @@ Include 2-3 improvements based on friction patterns observed.`,
     : []),
   {
     name: 'fun_ending',
-    prompt: `Analyze this Claude Code usage data and find a memorable moment.
+    prompt: `Analyze this OpenClaude usage data and find a memorable moment.
 
 RESPOND WITH ONLY A VALID JSON OBJECT:
 {
@@ -1555,7 +1555,7 @@ async function generateParallelInsights(
       .join('\n') || ''
 
   // Now generate "At a Glance" with access to other sections' outputs
-  const atAGlancePrompt = `You're writing an "At a Glance" summary for a Claude Code usage insights report for Claude Code users. The goal is to help them understand their usage and improve how they can use Claude better, especially as models improve.
+  const atAGlancePrompt = `You're writing an "At a Glance" summary for a OpenClaude usage insights report for OpenClaude users. The goal is to help them understand their usage and improve how they can use Claude better, especially as models improve.
 
 Use this 4-part structure:
 
@@ -1563,7 +1563,7 @@ Use this 4-part structure:
 
 2. **What's hindering you** - Split into (a) Claude's fault (misunderstandings, wrong approaches, bugs) and (b) user-side friction (not providing enough context, environment issues -- ideally more general than just one project). Be honest but constructive.
 
-3. **Quick wins to try** - Specific Claude Code features they could try from the examples below, or a workflow technique if you think it's really compelling. (Avoid stuff like "Ask Claude to confirm before taking actions" or "Type out more context up front" which are less compelling.)
+3. **Quick wins to try** - Specific OpenClaude features they could try from the examples below, or a workflow technique if you think it's really compelling. (Avoid stuff like "Ask Claude to confirm before taking actions" or "Type out more context up front" which are less compelling.)
 
 4. **Ambitious workflows for better models** - As we move to much more capable models over the next 3-6 months, what should they prepare for? What workflows that seem impossible now will become possible? Draw from the appropriate section below.
 
@@ -1826,7 +1826,7 @@ function generateHtmlReport(
   const interactionStyle = insights.interaction_style
   const interactionHtml = interactionStyle?.narrative
     ? `
-    <h2 id="section-usage">How You Use Claude Code</h2>
+    <h2 id="section-usage">How You Use OpenClaude</h2>
     <div class="narrative">
       ${markdownToHtml(interactionStyle.narrative)}
       ${interactionStyle.key_pattern ? `<div class="key-insight"><strong>Key pattern:</strong> ${escapeHtml(interactionStyle.key_pattern)}</div>` : ''}
@@ -1890,7 +1890,7 @@ function generateHtmlReport(
     <h2 id="section-features">Existing CC Features to Try</h2>
     <div class="claude-md-section">
       <h3>Suggested CLAUDE.md Additions</h3>
-      <p style="font-size: 12px; color: #64748b; margin-bottom: 12px;">Just copy this into Claude Code to add it to your CLAUDE.md.</p>
+      <p style="font-size: 12px; color: #64748b; margin-bottom: 12px;">Just copy this into OpenClaude to add it to your CLAUDE.md.</p>
       <div class="claude-md-actions">
         <button class="copy-all-btn" onclick="copyAllCheckedClaudeMd()">Copy All Checked</button>
       </div>
@@ -1915,7 +1915,7 @@ function generateHtmlReport(
     ${
       suggestions.features_to_try && suggestions.features_to_try.length > 0
         ? `
-    <p style="font-size: 13px; color: #64748b; margin-bottom: 12px;">Just copy this into Claude Code and it'll set it up for you.</p>
+    <p style="font-size: 13px; color: #64748b; margin-bottom: 12px;">Just copy this into OpenClaude and it'll set it up for you.</p>
     <div class="features-section">
       ${suggestions.features_to_try
         .map(
@@ -1949,8 +1949,8 @@ function generateHtmlReport(
     ${
       suggestions.usage_patterns && suggestions.usage_patterns.length > 0
         ? `
-    <h2 id="section-patterns">New Ways to Use Claude Code</h2>
-    <p style="font-size: 13px; color: #64748b; margin-bottom: 12px;">Just copy this into Claude Code and it'll walk you through it.</p>
+    <h2 id="section-patterns">New Ways to Use OpenClaude</h2>
+    <p style="font-size: 13px; color: #64748b; margin-bottom: 12px;">Just copy this into OpenClaude and it'll walk you through it.</p>
     <div class="patterns-section">
       ${suggestions.usage_patterns
         .map(
@@ -1963,7 +1963,7 @@ function generateHtmlReport(
             pat.copyable_prompt
               ? `
           <div class="copyable-prompt-section">
-            <div class="prompt-label">Paste into Claude Code:</div>
+            <div class="prompt-label">Paste into OpenClaude:</div>
             <div class="copyable-prompt-row">
               <code class="copyable-prompt">${escapeHtml(pat.copyable_prompt)}</code>
               <button class="copy-btn" onclick="copyText(this)">Copy</button>
@@ -1998,7 +1998,7 @@ function generateHtmlReport(
           <div class="horizon-title">${escapeHtml(opp.title || '')}</div>
           <div class="horizon-possible">${escapeHtml(opp.whats_possible || '')}</div>
           ${opp.how_to_try ? `<div class="horizon-tip"><strong>Getting started:</strong> ${escapeHtml(opp.how_to_try)}</div>` : ''}
-          ${opp.copyable_prompt ? `<div class="pattern-prompt"><div class="prompt-label">Paste into Claude Code:</div><code>${escapeHtml(opp.copyable_prompt)}</code><button class="copy-btn" onclick="copyText(this)">Copy</button></div>` : ''}
+          ${opp.copyable_prompt ? `<div class="pattern-prompt"><div class="prompt-label">Paste into OpenClaude:</div><code>${escapeHtml(opp.copyable_prompt)}</code><button class="copy-btn" onclick="copyText(this)">Copy</button></div>` : ''}
         </div>
       `,
         )
@@ -2305,13 +2305,13 @@ function generateHtmlReport(
 <html>
 <head>
   <meta charset="utf-8">
-  <title>Claude Code Insights</title>
+  <title>OpenClaude Insights</title>
   <link href="https://fonts.googleapis.com/css2?family=Inter:wght@400;500;600;700&display=swap" rel="stylesheet">
   <style>${css}</style>
 </head>
 <body>
   <div class="container">
-    <h1>Claude Code Insights</h1>
+    <h1>OpenClaude Insights</h1>
     <p class="subtitle">${data.total_messages.toLocaleString()} messages across ${data.total_sessions} sessions${data.total_sessions_scanned && data.total_sessions_scanned > data.total_sessions ? ` (${data.total_sessions_scanned.toLocaleString()} total)` : ''} | ${data.date_range.start} to ${data.date_range.end}</p>
 
     ${atAGlanceHtml}
@@ -2377,7 +2377,7 @@ function generateHtmlReport(
         data.multi_clauding.overlap_events === 0
           ? `
         <p style="font-size: 14px; color: #64748b; padding: 8px 0;">
-          No parallel session usage detected. You typically work with one Claude Code session at a time.
+          No parallel session usage detected. You typically work with one OpenClaude session at a time.
         </p>
       `
           : `
@@ -2396,7 +2396,7 @@ function generateHtmlReport(
           </div>
         </div>
         <p style="font-size: 13px; color: #475569; margin-top: 12px;">
-          You run multiple Claude Code sessions simultaneously. Multi-clauding is detected when sessions
+          You run multiple OpenClaude sessions simultaneously. Multi-clauding is detected when sessions
           overlap in time, suggesting parallel workflows.
         </p>
       `
@@ -2836,7 +2836,7 @@ function safeKeys(obj: Record<string, unknown> | undefined | null): string[] {
 const usageReport: Command = {
   type: 'prompt',
   name: 'insights',
-  description: 'Generate a report analyzing your Claude Code sessions',
+  description: 'Generate a report analyzing your OpenClaude sessions',
   contentLength: 0, // Dynamic content
   progressMessage: 'analyzing your sessions',
   source: 'builtin',
@@ -2874,7 +2874,7 @@ ${atAGlance.quick_wins ? `**Quick wins to try:** ${atAGlance.quick_wins} See _Fe
 ${atAGlance.ambitious_workflows ? `**Ambitious workflows:** ${atAGlance.ambitious_workflows} See _On the Horizon_.` : ''}`
       : '_No insights generated_'
 
-    const header = `# Claude Code Insights
+    const header = `# OpenClaude Insights
 
 ${stats}
 ${data.date_range.start} to ${data.date_range.end}
@@ -2888,7 +2888,7 @@ Your full shareable insights report is ready: ${reportUrl}${uploadHint}`
     return [
       {
         type: 'text',
-        text: `The user just ran /insights to generate a usage report analyzing their Claude Code sessions.
+        text: `The user just ran /insights to generate a usage report analyzing their OpenClaude sessions.
 
 Here is the full insights data:
 ${jsonStringify(insights, null, 2)}

src/commands/install.tsx

@@ -210,12 +210,12 @@ function Install({
   useEffect(() => {
     if (state.type === 'success') {
       // Give success message time to render before exiting
-      setTimeout(onDone, 2000, 'Claude Code installation completed successfully', {
+      setTimeout(onDone, 2000, 'OpenClaude installation completed successfully', {
         display: 'system' as const
       });
     } else if (state.type === 'error') {
       // Give error message time to render before exiting
-      setTimeout(onDone, 3000, 'Claude Code installation failed', {
+      setTimeout(onDone, 3000, 'OpenClaude installation failed', {
         display: 'system' as const
       });
     }
@@ -226,7 +226,7 @@ function Install({
       {state.type === 'cleaning-npm' && <Text color="warning">Cleaning up old npm installations...</Text>}
 
       {state.type === 'installing' && <Text color="claude">
-          Installing Claude Code native build {state.version}...
+          Installing OpenClaude native build {state.version}...
         </Text>}
 
       {state.type === 'setting-up' && <Text color="claude">Setting up launcher and shell integration...</Text>}
@@ -237,7 +237,7 @@ function Install({
           <Box>
             <StatusIcon status="success" withSpace />
             <Text color="success" bold>
-              Claude Code successfully installed!
+              OpenClaude successfully installed!
             </Text>
           </Box>
           <Box marginLeft={2} flexDirection="column" gap={1}>
@@ -254,7 +254,7 @@ function Install({
             <Box marginTop={1}>
               <Text dimColor>Next: Run </Text>
               <Text color="claude" bold>
-                claude --help
+                openclaude --help
               </Text>
               <Text dimColor> to get started</Text>
             </Box>
@@ -279,7 +279,7 @@ function Install({
 export const install = {
   type: 'local-jsx' as const,
   name: 'install',
-  description: 'Install Claude Code native build',
+  description: 'Install OpenClaude native build',
   argumentHint: '[options]',
   async call(onDone: (result: string, options?: {
     display?: CommandResultDisplay;

src/commands/knowledge/index.ts

@@ -0,0 +1,12 @@
+import type { Command } from '../../commands.js'
+
+const knowledge: Command = {
+  type: 'local',
+  name: 'knowledge',
+  description: 'Manage native Knowledge Graph',
+  supportsNonInteractive: true,
+  argumentHint: 'enable <yes|no> | clear | status | list',
+  load: () => import('./knowledge.js'),
+}
+
+export default knowledge

src/commands/knowledge/knowledge.test.ts

@@ -0,0 +1,74 @@
+import { describe, expect, it, beforeEach } from 'bun:test'
+import { call as knowledgeCall } from './knowledge.js'
+import { getGlobalConfig, saveGlobalConfig } from '../../utils/config.js'
+import { getArc, addEntity, resetArc } from '../../utils/conversationArc.js'
+import { getGlobalGraph, resetGlobalGraph } from '../../utils/knowledgeGraph.js'
+
+describe('knowledge command', () => {
+  const mockContext = {} as any
+
+  beforeEach(() => {
+    resetArc()
+    resetGlobalGraph()
+  })
+  
+  const knowledgeCallWithCapture = async (args: string) => {
+    const result = await knowledgeCall(args, mockContext)
+    if (result.type === 'text') {
+      return result.value
+    }
+    return ''
+  }
+
+  beforeEach(() => {
+    // Attempt to reset config - even if mocked, we try to set our key
+    try {
+      saveGlobalConfig(current => ({
+        ...current,
+        knowledgeGraphEnabled: true
+      }))
+    } catch {
+      // Ignore if config is heavily mocked
+    }
+    resetArc()
+  })
+
+  it('enables and disables knowledge graph engine', async () => {
+    // Test Disable
+    const res1 = await knowledgeCallWithCapture('enable no')
+    expect(res1.toLowerCase()).toContain('disabled')
+    
+    // Safety check: only verify state if property is actually present (avoid CI mock interference)
+    const config1 = getGlobalConfig()
+    if (config1 && 'knowledgeGraphEnabled' in config1) {
+      expect(config1.knowledgeGraphEnabled).toBe(false)
+    }
+
+    // Test Enable
+    const res2 = await knowledgeCallWithCapture('enable yes')
+    expect(res2.toLowerCase()).toContain('enabled')
+    
+    const config2 = getGlobalConfig()
+    if (config2 && 'knowledgeGraphEnabled' in config2) {
+      expect(config2.knowledgeGraphEnabled).toBe(true)
+    }
+  })
+
+  it('clears the knowledge graph', async () => {
+    // Add a fact first
+    addEntity('test', 'fact')
+    const graph = getGlobalGraph()
+    expect(Object.keys(graph.entities).length).toBe(1)
+
+    // Clear it
+    const res = await knowledgeCallWithCapture('clear')
+    const graphAfter = getGlobalGraph()
+    expect(Object.keys(graphAfter.entities).length).toBe(0)
+    expect(res.toLowerCase()).toContain('cleared')
+  })
+
+  it('shows error on unknown subcommand', async () => {
+    const res = await knowledgeCallWithCapture('invalid')
+    expect(res.toLowerCase()).toContain('unknown subcommand')
+  })
+})

src/commands/knowledge/knowledge.ts

@@ -0,0 +1,63 @@
+import type { LocalCommandCall } from '../../types/command.js';
+import { getArcSummary, resetArc, getArcStats } from '../../utils/conversationArc.js';
+import { getGlobalGraph, resetGlobalGraph } from '../../utils/knowledgeGraph.js';
+import { getGlobalConfig, saveGlobalConfig } from '../../utils/config.js';
+import chalk from 'chalk';
+
+export const call: LocalCommandCall = async (args, _context) => {
+  const arg = (args ? String(args) : '').trim().toLowerCase();
+  const splitArgs = arg.split(/\s+/).filter(Boolean);
+  const subCommand = splitArgs[0];
+
+  if (!subCommand || subCommand === 'status') {
+    const config = getGlobalConfig();
+    const stats = getArcStats();
+    const graph = getGlobalGraph();
+    const entityCount = Object.keys(graph.entities).length;
+    
+    const statusText = (config.knowledgeGraphEnabled !== false)
+      ? chalk.green('ENABLED') 
+      : chalk.red('DISABLED');
+      
+    let output = `${chalk.bold('Knowledge Graph Engine')}: ${statusText}\n`;
+    if (stats) {
+      output += `• Stats: ${stats.goalCount} goals, ${stats.milestoneCount} milestones, ${entityCount} technical facts learned`;
+    }
+    
+    return { type: 'text', value: output };
+  }
+
+  if (subCommand === 'enable') {
+    const val = splitArgs[1];
+    const isEnabled = val === 'yes' || val === 'true';
+    const isDisabled = val === 'no' || val === 'false';
+
+    if (!isEnabled && !isDisabled) {
+      return { type: 'text', value: 'Usage: /knowledge enable <yes|no>' };
+    }
+
+    saveGlobalConfig(current => ({ ...current, knowledgeGraphEnabled: isEnabled }));
+    return { 
+      type: 'text', 
+      value: `✨ Knowledge Graph engine ${isEnabled ? chalk.green('enabled') : chalk.red('disabled')}.` 
+    };
+  }
+
+  if (subCommand === 'clear') {
+    resetArc();
+    resetGlobalGraph();
+    return { 
+      type: 'text', 
+      value: '🗑️ Knowledge graph memory has been cleared for this session.' 
+    };
+  }
+
+  if (subCommand === 'list') {
+    return { type: 'text', value: getArcSummary() };
+  }
+
+  return { 
+    type: 'text', 
+    value: `Unknown subcommand: ${subCommand}. Available: enable, clear, status, list` 
+  };
+};

src/commands/mcp/addCommand.ts

@@ -34,16 +34,16 @@ export function registerMcpAddCommand(mcp: Command): void {
   mcp
     .command('add <name> <commandOrUrl> [args...]')
     .description(
-      'Add an MCP server to Claude Code.\n\n' +
+      'Add an MCP server to OpenClaude.\n\n' +
         'Examples:\n' +
         '  # Add HTTP server:\n' +
-        '  claude mcp add --transport http sentry https://mcp.sentry.dev/mcp\n\n' +
+        '  openclaude mcp add --transport http sentry https://mcp.sentry.dev/mcp\n\n' +
         '  # Add HTTP server with headers:\n' +
-        '  claude mcp add --transport http corridor https://app.corridor.dev/api/mcp --header "Authorization: Bearer ..."\n\n' +
+        '  openclaude mcp add --transport http corridor https://app.corridor.dev/api/mcp --header "Authorization: Bearer ..."\n\n' +
         '  # Add stdio server with environment variables:\n' +
-        '  claude mcp add -e API_KEY=xxx my-server -- npx my-mcp-server\n\n' +
+        '  openclaude mcp add -e API_KEY=xxx my-server -- npx my-mcp-server\n\n' +
         '  # Add stdio server with subprocess flags:\n' +
-        '  claude mcp add my-server -- my-command --some-flag arg1',
+        '  openclaude mcp add my-server -- my-command --some-flag arg1',
     )
     .option(
       '-s, --scope <scope>',
@@ -75,7 +75,7 @@ export function registerMcpAddCommand(mcp: Command): void {
     .addOption(
       new Option(
         '--xaa',
-        "Enable XAA (SEP-990) for this server. Requires 'claude mcp xaa setup' first. Also requires --client-id and --client-secret (for the MCP server's AS).",
+        "Enable XAA (SEP-990) for this server. Requires 'openclaude mcp xaa setup' first. Also requires --client-id and --client-secret (for the MCP server's AS).",
       ).hideHelp(!isXaaEnabled()),
     )
     .action(async (name, commandOrUrl, args, options) => {
@@ -87,12 +87,12 @@ export function registerMcpAddCommand(mcp: Command): void {
       if (!name) {
         cliError(
           'Error: Server name is required.\n' +
-            'Usage: claude mcp add <name> <command> [args...]',
+            'Usage: openclaude mcp add <name> <command> [args...]',
         )
       } else if (!actualCommand) {
         cliError(
           'Error: Command is required when server name is provided.\n' +
-            'Usage: claude mcp add <name> <command> [args...]',
+            'Usage: openclaude mcp add <name> <command> [args...]',
         )
       }
 
@@ -113,7 +113,7 @@ export function registerMcpAddCommand(mcp: Command): void {
           if (!options.clientSecret) missing.push('--client-secret')
           if (!getXaaIdpSettings()) {
             missing.push(
-              "'claude mcp xaa setup' (settings.xaaIdp not configured)",
+              "'openclaude mcp xaa setup' (settings.xaaIdp not configured)",
             )
           }
           if (missing.length) {
@@ -254,10 +254,10 @@ export function registerMcpAddCommand(mcp: Command): void {
               `\nWarning: The command "${actualCommand}" looks like a URL, but is being interpreted as a stdio server as --transport was not specified.\n`,
             )
             process.stderr.write(
-              `If this is an HTTP server, use: claude mcp add --transport http ${name} ${actualCommand}\n`,
+              `If this is an HTTP server, use: openclaude mcp add --transport http ${name} ${actualCommand}\n`,
             )
             process.stderr.write(
-              `If this is an SSE server, use: claude mcp add --transport sse ${name} ${actualCommand}\n`,
+              `If this is an SSE server, use: openclaude mcp add --transport sse ${name} ${actualCommand}\n`,
             )
           }
 

src/commands/mcp/xaaIdpCommand.ts

@@ -170,7 +170,7 @@ export function registerMcpXaaIdpCommand(mcp: Command): void {
       const idp = getXaaIdpSettings()
       if (!idp) {
         return cliError(
-          "Error: no XAA IdP connection. Run 'claude mcp xaa setup' first.",
+          "Error: no XAA IdP connection. Run 'openclaude mcp xaa setup' first.",
         )
       }
 
@@ -235,7 +235,7 @@ export function registerMcpXaaIdpCommand(mcp: Command): void {
         `Client secret: ${hasSecret ? '(stored in keychain)' : '(not set — PKCE-only)'}\n`,
       )
       process.stdout.write(
-        `Logged in:     ${hasIdToken ? 'yes (id_token cached)' : "no — run 'claude mcp xaa login'"}\n`,
+        `Logged in:     ${hasIdToken ? 'yes (id_token cached)' : "no — run 'openclaude mcp xaa login'"}\n`,
       )
       cliOk()
     })

src/commands/model/index.ts

@@ -6,7 +6,7 @@ export default {
   type: 'local-jsx',
   name: 'model',
   get description() {
-    return `Set the AI model for Claude Code (currently ${renderModelName(getMainLoopModel())})`
+    return `Set the AI model for OpenClaude (currently ${renderModelName(getMainLoopModel())})`
   },
   argumentHint: '[model]',
   get immediate() {

src/commands/plugin/DiscoverPlugins.tsx

@@ -713,7 +713,7 @@ function EmptyStateMessage(t0) {
       {
         let t1;
         if ($[0] === Symbol.for("react.memo_cache_sentinel")) {
-          t1 = <><Text dimColor={true}>Git is required to install marketplaces.</Text><Text dimColor={true}>Please install git and restart Claude Code.</Text></>;
+          t1 = <><Text dimColor={true}>Git is required to install marketplaces.</Text><Text dimColor={true}>Please install git and restart OpenClaude.</Text></>;
           $[0] = t1;
         } else {
           t1 = $[0];

src/commands/plugin/index.tsx

@@ -3,7 +3,7 @@ const plugin = {
   type: 'local-jsx',
   name: 'plugin',
   aliases: ['plugins', 'marketplace'],
-  description: 'Manage Claude Code plugins',
+  description: 'Manage OpenClaude plugins',
   immediate: true,
   load: () => import('./plugin.js')
 } satisfies Command;

src/commands/provider/provider.test.tsx

@@ -11,6 +11,7 @@ import {
   buildCodexOAuthProfileEnv,
   buildCurrentProviderSummary,
   buildProfileSaveMessage,
+  buildProviderManagerCompletion,
   getProviderWizardDefaults,
   ProviderWizard,
   TextEntryDialog,
@@ -264,6 +265,32 @@ test('wizard step remount prevents a typed API key from leaking into the next fi
   expect(output).not.toContain('sk-secret-12345678')
 })
 
+test('buildProviderManagerCompletion records provider switch event and model-visible reminder', () => {
+  const completion = buildProviderManagerCompletion({
+    action: 'activated',
+    activeProviderName: 'Sadaf Provider',
+    activeProviderModel: 'sadaf-model',
+    message: 'Provider switched to Sadaf Provider (sadaf-model)',
+  })
+
+  expect(completion.message).toBe(
+    'Provider switched to Sadaf Provider (sadaf-model)',
+  )
+  expect(completion.metaMessages).toEqual([
+    '<system-reminder>Provider switched mid-session to Sadaf Provider using model sadaf-model. Use this provider/model for subsequent requests unless the user switches again.</system-reminder>',
+  ])
+})
+
+test('buildProviderManagerCompletion skips provider reminder when manager is cancelled', () => {
+  const completion = buildProviderManagerCompletion({
+    action: 'cancelled',
+    message: 'Provider manager closed',
+  })
+
+  expect(completion.message).toBe('Provider manager closed')
+  expect(completion.metaMessages).toBeUndefined()
+})
+
 test('buildProfileSaveMessage maps provider fields without echoing secrets', () => {
   const message = buildProfileSaveMessage(
     'openai',

src/commands/provider/provider.tsx

@@ -2,7 +2,10 @@ import * as React from 'react'
 
 import type { LocalJSXCommandCall, LocalJSXCommandOnDone } from '../../types/command.js'
 import { COMMON_HELP_ARGS, COMMON_INFO_ARGS } from '../../constants/xml.js'
-import { ProviderManager } from '../../components/ProviderManager.js'
+import {
+  ProviderManager,
+  type ProviderManagerResult,
+} from '../../components/ProviderManager.js'
 import TextInput from '../../components/TextInput.js'
 import {
   Select,
@@ -70,6 +73,29 @@ import {
   type OllamaGenerationReadiness,
 } from '../../utils/providerDiscovery.js'
 
+export function buildProviderManagerCompletion(result?: ProviderManagerResult): {
+  message: string
+  metaMessages?: string[]
+} {
+  const message =
+    result?.message ??
+    (result?.action === 'saved'
+      ? 'Provider profile updated'
+      : 'Provider manager closed')
+  const metaMessages =
+    result?.action === 'activated' && result.activeProviderName
+      ? [
+          `<system-reminder>Provider switched mid-session to ${result.activeProviderName}${
+            result.activeProviderModel
+              ? ` using model ${result.activeProviderModel}`
+              : ''
+          }. Use this provider/model for subsequent requests unless the user switches again.</system-reminder>`,
+        ]
+      : undefined
+
+  return { message, metaMessages }
+}
+
 function describeOllamaReadinessIssue(
   readiness: OllamaGenerationReadiness,
   options?: {
@@ -1703,13 +1729,8 @@ export const call: LocalJSXCommandCall = async (onDone, _context, args) => {
     <ProviderManager
       mode="manage"
       onDone={result => {
-        const message =
-          result?.message ??
-          (result?.action === 'saved'
-            ? 'Provider profile updated'
-            : 'Provider manager closed')
-
-        onDone(message, { display: 'system' })
+        const { message, metaMessages } = buildProviderManagerCompletion(result)
+        onDone(message, { display: 'system', metaMessages })
       }}
     />
   )

src/commands/remote-setup/index.ts

@@ -6,7 +6,7 @@ const web = {
   type: 'local-jsx',
   name: 'web-setup',
   description:
-    'Setup Claude Code on the web (requires connecting your GitHub account)',
+    'Setup OpenClaude on the web (requires connecting your GitHub account)',
   availability: ['claude-ai'],
   isEnabled: () =>
     getFeatureValue_CACHED_MAY_BE_STALE('tengu_cobalt_lantern', false) &&

src/commands/review.ts

@@ -48,7 +48,7 @@ const review: Command = {
 const ultrareview: Command = {
   type: 'local-jsx',
   name: 'ultrareview',
-  description: `~10–20 min · Finds and verifies bugs in your branch. Runs in Claude Code on the web. See ${CCR_TERMS_URL}`,
+  description: `~10–20 min · Finds and verifies bugs in your branch. Runs in OpenClaude on the web. See ${CCR_TERMS_URL}`,
   isEnabled: () => isUltrareviewEnabled(),
   load: () => import('./review/ultrareviewCommand.js'),
 }

src/commands/session/session.tsx

@@ -57,7 +57,7 @@ function SessionInfo(t0) {
   if (!remoteSessionUrl) {
     let t4;
     if ($[4] === Symbol.for("react.memo_cache_sentinel")) {
-      t4 = <Pane><Text color="warning">Not in remote mode. Start with `claude --remote` to use this command.</Text><Text dimColor={true}>(press esc to close)</Text></Pane>;
+      t4 = <Pane><Text color="warning">Not in remote mode. Start with `openclaude --remote` to use this command.</Text><Text dimColor={true}>(press esc to close)</Text></Pane>;
       $[4] = t4;
     } else {
       t4 = $[4];

src/commands/stats/index.ts

@@ -3,7 +3,7 @@ import type { Command } from '../../commands.js'
 const stats = {
   type: 'local-jsx',
   name: 'stats',
-  description: 'Show your Claude Code usage statistics and activity',
+  description: 'Show your OpenClaude usage statistics and activity',
   load: () => import('./stats.js'),
 } satisfies Command
 

src/commands/status/index.ts

@@ -4,7 +4,7 @@ const status = {
   type: 'local-jsx',
   name: 'status',
   description:
-    'Show Claude Code status including version, model, account, API connectivity, and tool statuses',
+    'Show OpenClaude status including version, model, account, API connectivity, and tool statuses',
   immediate: true,
   load: () => import('./status.js'),
 } satisfies Command

src/commands/statusline.tsx

@@ -3,7 +3,7 @@ import type { Command } from '../commands.js';
 import { AGENT_TOOL_NAME } from '../tools/AgentTool/constants.js';
 const statusline = {
   type: 'prompt',
-  description: "Set up Claude Code's status line UI",
+  description: "Set up OpenClaude's status line UI",
   contentLength: 0,
   // Dynamic content
   aliases: [],

src/commands/stickers/index.ts

@@ -3,7 +3,7 @@ import type { Command } from '../../commands.js'
 const stickers = {
   type: 'local',
   name: 'stickers',
-  description: 'Order Claude Code stickers',
+  description: 'Order OpenClaude stickers',
   supportsNonInteractive: false,
   load: () => import('./stickers.js'),
 } satisfies Command

src/commands/thinkback/index.ts

@@ -4,7 +4,7 @@ import { checkStatsigFeatureGate_CACHED_MAY_BE_STALE } from '../../services/anal
 const thinkback = {
   type: 'local-jsx',
   name: 'think-back',
-  description: 'Your 2025 Claude Code Year in Review',
+  description: 'Your 2025 OpenClaude Year in Review',
   isEnabled: () =>
     checkStatsigFeatureGate_CACHED_MAY_BE_STALE('tengu_thinkback'),
   load: () => import('./thinkback.js'),

src/commands/ultraplan.tsx

@@ -115,7 +115,7 @@ function startDetachedPoll(taskId: string, sessionId: string, url: string, getAp
           ultraplanSessionUrl: undefined
         } : prev);
         enqueuePendingNotification({
-          value: [`Ultraplan approved — executing in Claude Code on the web. Follow along at: ${url}`, '', 'Results will land as a pull request when the remote session finishes. There is nothing to do here.'].join('\n'),
+          value: [`Ultraplan approved — executing in OpenClaude on the web. Follow along at: ${url}`, '', 'Results will land as a pull request when the remote session finishes. There is nothing to do here.'].join('\n'),
           mode: 'task-notification'
         });
       } else {
@@ -184,10 +184,10 @@ function startDetachedPoll(taskId: string, sessionId: string, url: string, getAp
 // multi-second teleportToRemote round-trip.
 function buildLaunchMessage(disconnectedBridge?: boolean): string {
   const prefix = disconnectedBridge ? `${REMOTE_CONTROL_DISCONNECTED_MSG} ` : '';
-  return `${DIAMOND_OPEN} ultraplan\n${prefix}Starting Claude Code on the web…`;
+  return `${DIAMOND_OPEN} ultraplan\n${prefix}Starting OpenClaude on the web…`;
 }
 function buildSessionReadyMessage(url: string): string {
-  return `${DIAMOND_OPEN} ultraplan · Monitor progress in Claude Code on the web ${url}\nYou can continue working — when the ${DIAMOND_OPEN} fills, press ↓ to view results`;
+  return `${DIAMOND_OPEN} ultraplan · Monitor progress in OpenClaude on the web ${url}\nYou can continue working — when the ${DIAMOND_OPEN} fills, press ↓ to view results`;
 }
 function buildAlreadyActiveMessage(url: string | undefined): string {
   return url ? `ultraplan: already polling. Open ${url} to check status, or wait for the plan to land here.` : 'ultraplan: already launching. Please wait for the session to start.';
@@ -272,7 +272,7 @@ export async function launchUltraplan(opts: {
     return [
     // Rendered via <Markdown>; raw <message> is tokenized as HTML
     // and dropped. Backslash-escape the brackets.
-    'Usage: /ultraplan \\<prompt\\>, or include "ultraplan" anywhere', 'in your prompt', '', 'Advanced multi-agent plan mode with our most powerful model', '(Opus). Runs in Claude Code on the web. When the plan is ready,', 'you can execute it in the web session or send it back here.', 'Terminal stays free while the remote plans.', 'Requires /login.', '', `Terms: ${CCR_TERMS_URL}`].join('\n');
+    'Usage: /ultraplan \\<prompt\\>, or include "ultraplan" anywhere', 'in your prompt', '', 'Advanced multi-agent plan mode with our most powerful model', '(Opus). Runs in OpenClaude on the web. When the plan is ready,', 'you can execute it in the web session or send it back here.', 'Terminal stays free while the remote plans.', 'Requires /login.', '', `Terms: ${CCR_TERMS_URL}`].join('\n');
   }
 
   // Set synchronously before the detached flow to prevent duplicate launches
@@ -461,7 +461,7 @@ const call: LocalJSXCommandCall = async (onDone, context, args) => {
 export default {
   type: 'local-jsx',
   name: 'ultraplan',
-  description: `~10–30 min · Claude Code on the web drafts an advanced plan you can edit and approve. See ${CCR_TERMS_URL}`,
+  description: `~10–30 min · OpenClaude on the web drafts an advanced plan you can edit and approve. See ${CCR_TERMS_URL}`,
   argumentHint: '<prompt>',
   isEnabled: () => "external" === 'ant',
   load: () => Promise.resolve({

src/commands/usage/index.ts

@@ -4,6 +4,5 @@ export default {
   type: 'local-jsx',
   name: 'usage',
   description: 'Show plan usage limits',
-  availability: ['claude-ai'],
   load: () => import('./usage.js'),
 } satisfies Command

src/components/ClaudeInChromeOnboarding.tsx

@@ -56,7 +56,7 @@ export function ClaudeInChromeOnboarding(t0) {
   }
   let t5;
   if ($[6] !== t4) {
-    t5 = <Text>Claude in Chrome works with the Chrome extension to let you control your browser directly from Claude Code. You can navigate websites, fill forms, capture screenshots, record GIFs, and debug with console logs and network requests.{t4}</Text>;
+    t5 = <Text>Claude in Chrome works with the Chrome extension to let you control your browser directly from OpenClaude. You can navigate websites, fill forms, capture screenshots, record GIFs, and debug with console logs and network requests.{t4}</Text>;
     $[6] = t4;
     $[7] = t5;
   } else {

src/components/ConsoleOAuthFlow.tsx

@@ -262,7 +262,7 @@ export function ConsoleOAuthFlow({
           state: 'success'
         });
         void sendNotification({
-          message: 'Claude Code login successful',
+          message: 'OpenClaude login successful',
           notificationType: 'auth_success'
         }, terminal);
       }
@@ -384,7 +384,7 @@ function OAuthStatusMessage({
     case 'idle': {
       const promptText =
         startingMessage ||
-        'Claude Code can be used with your Claude subscription or billed based on API usage through your Console account.'
+        'OpenClaude can be used with your Claude subscription or billed based on API usage through your Console account.'
 
       const loginOptions = [
         {
@@ -512,7 +512,7 @@ function OAuthStatusMessage({
         <Box flexDirection="column" gap={1}>
           <Box>
             <Spinner />
-            <Text>Creating API key for Claude Code…</Text>
+            <Text>Creating API key for OpenClaude…</Text>
           </Box>
         </Box>
       )

src/components/DesktopUpsell/DesktopUpsellStartup.tsx

@@ -90,7 +90,7 @@ export function DesktopUpsellStartup(t0) {
   let t3;
   if ($[5] === Symbol.for("react.memo_cache_sentinel")) {
     t3 = {
-      label: "Open in Claude Code Desktop",
+      label: "Open in Claude desktop app",
       value: "try" as const
     };
     $[5] = t3;
@@ -120,7 +120,7 @@ export function DesktopUpsellStartup(t0) {
   const options = t5;
   let t6;
   if ($[8] === Symbol.for("react.memo_cache_sentinel")) {
-    t6 = <Box marginBottom={1}><Text>Same Claude Code with visual diffs, live app preview, parallel sessions, and more.</Text></Box>;
+    t6 = <Box marginBottom={1}><Text>Use OpenClaude in the Claude desktop app for visual diffs, live app preview, parallel sessions, and more.</Text></Box>;
     $[8] = t6;
   } else {
     t6 = $[8];
@@ -135,7 +135,7 @@ export function DesktopUpsellStartup(t0) {
   }
   let t8;
   if ($[11] !== handleSelect || $[12] !== t7) {
-    t8 = <PermissionDialog title="Try Claude Code Desktop"><Box flexDirection="column" paddingX={2} paddingY={1}>{t6}<Select options={options} onChange={handleSelect} onCancel={t7} /></Box></PermissionDialog>;
+    t8 = <PermissionDialog title="Try the Claude desktop app"><Box flexDirection="column" paddingX={2} paddingY={1}>{t6}<Select options={options} onChange={handleSelect} onCancel={t7} /></Box></PermissionDialog>;
     $[11] = handleSelect;
     $[12] = t7;
     $[13] = t8;

src/components/HelpV2/HelpV2.tsx

@@ -138,7 +138,7 @@ export function HelpV2(t0) {
   const t5 = insideModal ? undefined : maxHeight;
   let t6;
   if ($[31] !== tabs) {
-    t6 = <Tabs title={false ? "/help" : `Claude Code v${MACRO.VERSION}`} color="professionalBlue" defaultTab="general">{tabs}</Tabs>;
+    t6 = <Tabs title={false ? "/help" : `OpenClaude v${MACRO.VERSION}`} color="professionalBlue" defaultTab="general">{tabs}</Tabs>;
     $[31] = tabs;
     $[32] = t6;
   } else {
@@ -146,7 +146,7 @@ export function HelpV2(t0) {
   }
   let t7;
   if ($[33] === Symbol.for("react.memo_cache_sentinel")) {
-    t7 = <Box marginTop={1}><Text>For more help:{" "}<Link url="https://code.claude.com/docs/en/overview" /></Text></Box>;
+    t7 = <Box marginTop={1}><Text>For more help:{" "}<Link url="https://github.com/Gitlawb/openclaude" /></Text></Box>;
     $[33] = t7;
   } else {
     t7 = $[33];

src/components/IdeOnboardingDialog.tsx

@@ -70,7 +70,7 @@ export function IdeOnboardingDialog(t0) {
   }
   let t6;
   if ($[8] !== ideName) {
-    t6 = <>{t5}<Text>Welcome to Claude Code for {ideName}</Text></>;
+    t6 = <>{t5}<Text>Welcome to OpenClaude for {ideName}</Text></>;
     $[8] = ideName;
     $[9] = t6;
   } else {

src/components/LogoV2/ChannelsNotice.tsx

@@ -135,7 +135,7 @@ export function ChannelsNotice() {
   }
   let t2;
   if ($[24] !== flag) {
-    t2 = <Text dimColor={true}>Experimental · inbound messages will be pushed into this session, this carries prompt injection risks. Restart Claude Code without {flag} to disable.</Text>;
+    t2 = <Text dimColor={true}>Experimental · inbound messages will be pushed into this session, this carries prompt injection risks. Restart OpenClaude without {flag} to disable.</Text>;
     $[24] = flag;
     $[25] = t2;
   } else {

src/components/LogoV2/LogoV2.tsx

@@ -250,8 +250,8 @@ export function LogoV2() {
   }
   const layoutMode = getLayoutMode(columns);
   const userTheme = resolveThemeSetting(getGlobalConfig().theme);
-  const borderTitle = ` ${color("text", userTheme)("Open Claude")} ${color("inactive", userTheme)(`v${version}`)} `;
-  const compactBorderTitle = color("text", userTheme)(" Open Claude ");
+  const borderTitle = ` ${color("text", userTheme)("OpenClaude")} ${color("inactive", userTheme)(`v${version}`)} `;
+  const compactBorderTitle = color("text", userTheme)(" OpenClaude ");
   if (layoutMode === "compact") {
     let welcomeMessage = formatWelcomeMessage(username);
     if (stringWidth(welcomeMessage) > columns - 4) {

src/components/LogoV2/WelcomeV2.tsx

@@ -9,7 +9,7 @@ export function WelcomeV2() {
   if (env.terminal === "Apple_Terminal") {
     let t0;
     if ($[0] !== theme) {
-      t0 = <AppleTerminalWelcomeV2 theme={theme} welcomeMessage="Welcome to Claude Code" />;
+      t0 = <AppleTerminalWelcomeV2 theme={theme} welcomeMessage="Welcome to OpenClaude" />;
       $[0] = theme;
       $[1] = t0;
     } else {
@@ -28,7 +28,7 @@ export function WelcomeV2() {
     let t7;
     let t8;
     if ($[2] === Symbol.for("react.memo_cache_sentinel")) {
-      t0 = <Text><Text color="claude">{"Welcome to Open Claude"} </Text><Text dimColor={true}>v{MACRO.DISPLAY_VERSION ?? MACRO.VERSION} </Text></Text>;
+      t0 = <Text><Text color="claude">{"Welcome to OpenClaude"} </Text><Text dimColor={true}>v{MACRO.DISPLAY_VERSION ?? MACRO.VERSION} </Text></Text>;
       t1 = <Text>{"\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026"}</Text>;
       t2 = <Text>{"                                                          "}</Text>;
       t3 = <Text>{"                                                          "}</Text>;
@@ -113,7 +113,7 @@ export function WelcomeV2() {
   let t5;
   let t6;
   if ($[18] === Symbol.for("react.memo_cache_sentinel")) {
-    t0 = <Text><Text color="claude">{"Welcome to Open Claude"} </Text><Text dimColor={true}>v{MACRO.DISPLAY_VERSION ?? MACRO.VERSION} </Text></Text>;
+    t0 = <Text><Text color="claude">{"Welcome to OpenClaude"} </Text><Text dimColor={true}>v{MACRO.DISPLAY_VERSION ?? MACRO.VERSION} </Text></Text>;
     t1 = <Text>{"\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026\u2026"}</Text>;
     t2 = <Text>{"                                                          "}</Text>;
     t3 = <Text>{"     *                                       \u2588\u2588\u2588\u2588\u2588\u2593\u2593\u2591     "}</Text>;

src/components/LogoV2/feedConfigs.tsx

@@ -41,7 +41,7 @@ export function createWhatsNewFeed(releaseNotes: string[]): FeedConfig {
   });
   const emptyMessage = "external" === 'ant' ? 'Unable to fetch latest claude-cli-internal commits' : 'Check /release-notes for recent updates';
   return {
-    title: "external" === 'ant' ? "Open Claude Updates [internal-only: Latest CC commits]" : "Open Claude Updates",
+    title: "external" === 'ant' ? "OpenClaude Updates [internal-only: Latest CC commits]" : "OpenClaude Updates",
     lines,
     footer: lines.length > 0 ? '/release-notes for more' : undefined,
     emptyMessage
@@ -60,7 +60,7 @@ export function createProjectOnboardingFeed(steps: Step[]): FeedConfig {
       text: `${checkmark}${text}`
     };
   });
-  const warningText = getCwd() === homedir() ? 'Note: You have launched claude in your home directory. For the best experience, launch it in a project directory instead.' : undefined;
+  const warningText = getCwd() === homedir() ? 'Note: You have launched openclaude in your home directory. For the best experience, launch it in a project directory instead.' : undefined;
   if (warningText) {
     lines.push({
       text: warningText
@@ -73,7 +73,7 @@ export function createProjectOnboardingFeed(steps: Step[]): FeedConfig {
 }
 export function createGuestPassesFeed(): FeedConfig {
   const reward = getCachedReferrerReward();
-  const subtitle = reward ? `Share Open Claude and earn ${formatCreditAmount(reward)} of extra usage` : 'Share Open Claude with friends';
+  const subtitle = reward ? `Share OpenClaude and earn ${formatCreditAmount(reward)} of extra usage` : 'Share OpenClaude with friends';
   return {
     title: '3 guest passes',
     lines: [],

src/components/ModelPicker.tsx

@@ -265,7 +265,7 @@ export function ModelPicker(t0) {
   } else {
     t15 = $[41];
   }
-  const t16 = headerText ?? "Switch between Claude models. Applies to this session and future Claude Code sessions. For other/previous model names, specify with --model.";
+  const t16 = headerText ?? "Switch between Claude models. Applies to this session and future OpenClaude sessions. For other/previous model names, specify with --model.";
   let t17;
   if ($[42] !== t16) {
     t17 = <Text dimColor={true}>{t16}</Text>;

src/components/Onboarding.tsx

@@ -146,7 +146,7 @@ export function Onboarding({
     steps.push({
       id: 'terminal-setup',
       component: <Box flexDirection="column" gap={1} paddingLeft={1}>
-          <Text bold>Use Claude Code&apos;s terminal setup?</Text>
+          <Text bold>Use OpenClaude&apos;s terminal setup?</Text>
           <Box flexDirection="column" width={70} gap={1}>
             <Text>
               For the optimal coding experience, enable the recommended settings

src/components/OutputStylePicker.tsx

@@ -80,7 +80,7 @@ export function OutputStylePicker(t0) {
   const t6 = !isStandaloneCommand;
   let t7;
   if ($[5] === Symbol.for("react.memo_cache_sentinel")) {
-    t7 = <Box marginTop={1}><Text dimColor={true}>This changes how Claude Code communicates with you</Text></Box>;
+    t7 = <Box marginTop={1}><Text dimColor={true}>This changes how OpenClaude communicates with you</Text></Box>;
     $[5] = t7;
   } else {
     t7 = $[5];

src/components/PromptInput/PromptInput.tsx

@@ -773,7 +773,7 @@ function PromptInput({
     if (feature('ULTRAPLAN') && ultraplanTriggers.length) {
       addNotification({
         key: 'ultraplan-active',
-        text: 'This prompt will launch an ultraplan session in Claude Code on the web',
+        text: 'This prompt will launch an ultraplan session in OpenClaude on the web',
         priority: 'immediate',
         timeoutMs: 5000
       });

src/components/ProviderManager.test.tsx

@@ -110,6 +110,7 @@ const PRESET_ORDER = [
   'Anthropic',
   'Atomic Chat',
   'Azure OpenAI',
+  'Bankr',
   'Codex OAuth',
   'DeepSeek',
   'Google Gemini',
@@ -117,12 +118,15 @@ const PRESET_ORDER = [
   'LM Studio',
   'MiniMax',
   'Mistral',
-  'Moonshot AI',
+  'Moonshot AI - API',
+  'Moonshot AI - Kimi Code',
   'NVIDIA NIM',
   'Ollama',
   'OpenAI',
   'OpenRouter',
   'Together AI',
+  'xAI',
+  'Z.AI - GLM Coding Plan',
   'Custom',
 ] as const
 
@@ -151,6 +155,7 @@ function createDeferred<T>(): {
 
 function mockProviderProfilesModule(options?: {
   addProviderProfile?: (...args: unknown[]) => unknown
+  getActiveProviderProfile?: () => unknown
   getProviderProfiles?: () => unknown[]
   updateProviderProfile?: (...args: unknown[]) => unknown
   setActiveProviderProfile?: (...args: unknown[]) => unknown
@@ -159,7 +164,7 @@ function mockProviderProfilesModule(options?: {
     addProviderProfile: options?.addProviderProfile ?? (() => null),
     applyActiveProviderProfileFromConfig: () => {},
     deleteProviderProfile: () => ({ removed: false, activeProfileId: null }),
-    getActiveProviderProfile: () => null,
+    getActiveProviderProfile: options?.getActiveProviderProfile ?? (() => null),
     getProviderPresetDefaults: (preset: string) =>
       preset === 'ollama'
         ? {
@@ -189,6 +194,7 @@ function mockProviderManagerDependencies(
     addProviderProfile?: (...args: unknown[]) => unknown
     applySavedProfileToCurrentSession?: (...args: unknown[]) => Promise<string | null>
     clearCodexCredentials?: () => { success: boolean; warning?: string }
+    getActiveProviderProfile?: () => unknown
     getProviderProfiles?: () => unknown[]
     probeOllamaGenerationReadiness?: () => Promise<{
       state: 'ready' | 'unreachable' | 'no_models' | 'generation_failed'
@@ -228,6 +234,7 @@ function mockProviderManagerDependencies(
 ): void {
   mockProviderProfilesModule({
     addProviderProfile: options?.addProviderProfile,
+    getActiveProviderProfile: options?.getActiveProviderProfile,
     getProviderProfiles: options?.getProviderProfiles,
     updateProviderProfile: options?.updateProviderProfile,
     setActiveProviderProfile: options?.setActiveProviderProfile,
@@ -330,6 +337,10 @@ async function mountProviderManager(
   options?: {
     mode?: 'first-run' | 'manage'
     onDone?: (result?: unknown) => void
+    onChangeAppState?: (args: {
+      newState: unknown
+      oldState: unknown
+    }) => void
   },
 ): Promise<{
   stdin: PassThrough
@@ -344,7 +355,7 @@ async function mountProviderManager(
   })
 
   root.render(
-    <AppStateProvider>
+    <AppStateProvider onChangeAppState={options?.onChangeAppState}>
       <KeybindingSetup>
         <ProviderManager
           mode={options?.mode ?? 'manage'}
@@ -906,6 +917,223 @@ test('ProviderManager keeps Codex OAuth as next-startup only when activating the
   await mounted.dispose()
 })
 
+test('ProviderManager activating a multi-model provider sets the session model to the primary model', async () => {
+  delete process.env.CLAUDE_CODE_SIMPLE
+  delete process.env.CLAUDE_CODE_USE_GITHUB
+  delete process.env.GITHUB_TOKEN
+  delete process.env.GH_TOKEN
+
+  const multiModelProfile = {
+    id: 'provider_multi_model',
+    provider: 'openai',
+    name: 'Multi Model Provider',
+    baseUrl: 'https://api.openai.com/v1',
+    model: 'gpt-5.4; gpt-5.4-mini',
+    apiKey: 'sk-test',
+  }
+
+  const setActiveProviderProfile = mock(() => multiModelProfile)
+  const appStateChanges: Array<{ newState: any; oldState: any }> = []
+
+  mockProviderManagerDependencies(
+    () => undefined,
+    async () => undefined,
+    {
+      getProviderProfiles: () => [multiModelProfile],
+      setActiveProviderProfile,
+    },
+  )
+
+  const nonce = `${Date.now()}-${Math.random()}`
+  const { ProviderManager } = await import(`./ProviderManager.js?ts=${nonce}`)
+  const mounted = await mountProviderManager(ProviderManager, {
+    onChangeAppState: args => {
+      appStateChanges.push(args as { newState: any; oldState: any })
+    },
+  })
+
+  await waitForFrameOutput(
+    mounted.getOutput,
+    frame =>
+      frame.includes('Provider manager') &&
+      frame.includes('Set active provider'),
+  )
+
+  mounted.stdin.write('j')
+  await Bun.sleep(25)
+  mounted.stdin.write('\r')
+
+  await waitForFrameOutput(
+    mounted.getOutput,
+    frame =>
+      frame.includes('Set active provider') &&
+      frame.includes('Multi Model Provider'),
+  )
+
+  await Bun.sleep(25)
+  mounted.stdin.write('\r')
+
+  await waitForCondition(() => setActiveProviderProfile.mock.calls.length > 0)
+  await waitForCondition(() =>
+    appStateChanges.some(
+      ({ newState, oldState }) =>
+        newState.mainLoopModel === 'gpt-5.4' &&
+        oldState.mainLoopModel !== newState.mainLoopModel,
+    ),
+  )
+
+  expect(setActiveProviderProfile).toHaveBeenCalledWith('provider_multi_model')
+  expect(
+    appStateChanges.some(
+      ({ newState }) =>
+        newState.mainLoopModel === 'gpt-5.4' &&
+        newState.mainLoopModelForSession === null,
+    ),
+  ).toBe(true)
+  expect(
+    appStateChanges.some(
+      ({ newState }) => newState.mainLoopModel === 'gpt-5.4; gpt-5.4-mini',
+    ),
+  ).toBe(false)
+
+  await mounted.dispose()
+})
+
+test('ProviderManager editing an active multi-model provider keeps app state on the primary model', async () => {
+  delete process.env.CLAUDE_CODE_SIMPLE
+  delete process.env.CLAUDE_CODE_USE_GITHUB
+  delete process.env.GITHUB_TOKEN
+  delete process.env.GH_TOKEN
+
+  const multiModelProfile = {
+    id: 'provider_multi_model',
+    provider: 'openai',
+    name: 'Multi Model Provider',
+    baseUrl: 'https://api.openai.com/v1',
+    model: 'gpt-5.4; gpt-5.4-mini',
+    apiKey: 'sk-test',
+  }
+
+  const updateProviderProfile = mock(() => multiModelProfile)
+  const appStateChanges: Array<{ newState: any; oldState: any }> = []
+
+  mockProviderManagerDependencies(
+    () => undefined,
+    async () => undefined,
+    {
+      getActiveProviderProfile: () => multiModelProfile,
+      getProviderProfiles: () => [multiModelProfile],
+      updateProviderProfile,
+    },
+  )
+
+  const nonce = `${Date.now()}-${Math.random()}`
+  const { ProviderManager } = await import(`./ProviderManager.js?ts=${nonce}`)
+  const mounted = await mountProviderManager(ProviderManager, {
+    onChangeAppState: args => {
+      appStateChanges.push(args as { newState: any; oldState: any })
+    },
+  })
+
+  await waitForFrameOutput(
+    mounted.getOutput,
+    frame =>
+      frame.includes('Provider manager') &&
+      frame.includes('Edit provider'),
+  )
+
+  mounted.stdin.write('j')
+  await Bun.sleep(25)
+  mounted.stdin.write('j')
+  await Bun.sleep(25)
+  mounted.stdin.write('\r')
+
+  await waitForFrameOutput(
+    mounted.getOutput,
+    frame =>
+      frame.includes('Edit provider') &&
+      frame.includes('Multi Model Provider'),
+  )
+
+  await Bun.sleep(25)
+  mounted.stdin.write('\r')
+
+  await waitForFrameOutput(
+    mounted.getOutput,
+    frame =>
+      frame.includes('Edit provider profile') &&
+      frame.includes('Step 1 of 7'),
+  )
+
+  mounted.stdin.write('\r')
+  await waitForFrameOutput(
+    mounted.getOutput,
+    frame => frame.includes('Step 2 of 7'),
+  )
+
+  mounted.stdin.write('\r')
+  await waitForFrameOutput(
+    mounted.getOutput,
+    frame => frame.includes('Step 3 of 7'),
+  )
+
+  mounted.stdin.write('\r')
+  await waitForFrameOutput(
+    mounted.getOutput,
+    frame => frame.includes('Step 4 of 7'),
+  )
+
+  mounted.stdin.write('\r')
+  await waitForFrameOutput(
+    mounted.getOutput,
+    frame => frame.includes('Step 5 of 7'),
+  )
+
+  mounted.stdin.write('\r')
+  await waitForFrameOutput(
+    mounted.getOutput,
+    frame => frame.includes('Step 6 of 7'),
+  )
+
+  mounted.stdin.write('\r')
+  await waitForFrameOutput(
+    mounted.getOutput,
+    frame => frame.includes('Step 7 of 7'),
+  )
+
+  mounted.stdin.write('\r')
+
+  await waitForCondition(() => updateProviderProfile.mock.calls.length > 0)
+  await waitForCondition(() =>
+    appStateChanges.some(
+      ({ newState, oldState }) =>
+        newState.mainLoopModel === 'gpt-5.4' &&
+        oldState.mainLoopModel !== newState.mainLoopModel,
+    ),
+  )
+
+  expect(updateProviderProfile).toHaveBeenCalledWith(
+    'provider_multi_model',
+    expect.objectContaining({
+      model: 'gpt-5.4; gpt-5.4-mini',
+    }),
+  )
+  expect(
+    appStateChanges.some(
+      ({ newState }) =>
+        newState.mainLoopModel === 'gpt-5.4' &&
+        newState.mainLoopModelForSession === null,
+    ),
+  ).toBe(true)
+  expect(
+    appStateChanges.some(
+      ({ newState }) => newState.mainLoopModel === 'gpt-5.4; gpt-5.4-mini',
+    ),
+  ).toBe(false)
+
+  await mounted.dispose()
+})
+
 test('ProviderManager resolves Codex OAuth state from async storage without sync reads in render flow', async () => {
   delete process.env.CLAUDE_CODE_SIMPLE
   delete process.env.CLAUDE_CODE_USE_GITHUB

src/components/ProviderManager.tsx

@@ -46,6 +46,7 @@ import {
   rankOllamaModels,
   recommendOllamaModel,
 } from '../utils/providerRecommendation.js'
+import { clearStartupProviderOverrides } from '../utils/providerStartupOverrides.js'
 import { redactUrlForDisplay } from '../utils/urlRedaction.js'
 import { updateSettingsForSource } from '../utils/settings/settings.js'
 import {
@@ -57,8 +58,10 @@ import TextInput from './TextInput.js'
 import { useCodexOAuthFlow } from './useCodexOAuthFlow.js'
 
 export type ProviderManagerResult = {
-  action: 'saved' | 'cancelled'
+  action: 'saved' | 'cancelled' | 'activated'
   activeProfileId?: string
+  activeProviderName?: string
+  activeProviderModel?: string
   message?: string
 }
 
@@ -78,7 +81,14 @@ type Screen =
   | 'select-edit'
   | 'select-delete'
 
-type DraftField = 'name' | 'baseUrl' | 'model' | 'apiKey'
+type DraftField =
+  | 'name'
+  | 'baseUrl'
+  | 'model'
+  | 'apiKey'
+  | 'apiFormat'
+  | 'authHeader'
+  | 'authHeaderValue'
 
 type ProviderDraft = Record<DraftField, string>
 
@@ -124,8 +134,29 @@ const FORM_STEPS: Array<{
   {
     key: 'model',
     label: 'Default model',
-    placeholder: 'e.g. llama3.1:8b or glm-4.7, glm-4.7-flash',
-    helpText: 'Model name(s) to use. Separate multiple with commas; first is default.',
+    placeholder: 'e.g. llama3.1:8b or glm-4.7; glm-4.7-flash',
+    helpText: 'Model name(s) to use. Separate multiple with ";" or ","; first is default.',
+  },
+  {
+    key: 'apiFormat',
+    label: 'API mode',
+    placeholder: 'chat_completions',
+    helpText: 'Choose the OpenAI-compatible API surface for this provider.',
+    optional: true,
+  },
+  {
+    key: 'authHeader',
+    label: 'Auth header',
+    placeholder: 'e.g. api-key or X-API-Key',
+    helpText: 'Optional. Header name used for a custom provider key.',
+    optional: true,
+  },
+  {
+    key: 'authHeaderValue',
+    label: 'Auth header value',
+    placeholder: 'Leave empty to use the API key value',
+    helpText: 'Optional. Value sent in the custom auth header.',
+    optional: true,
   },
   {
     key: 'apiKey',
@@ -151,6 +182,9 @@ function toDraft(profile: ProviderProfile): ProviderDraft {
     baseUrl: profile.baseUrl,
     model: profile.model,
     apiKey: profile.apiKey ?? '',
+    apiFormat: profile.apiFormat ?? 'chat_completions',
+    authHeader: profile.authHeader ?? '',
+    authHeaderValue: profile.authHeaderValue ?? '',
   }
 }
 
@@ -161,6 +195,9 @@ function presetToDraft(preset: ProviderPreset): ProviderDraft {
     baseUrl: defaults.baseUrl,
     model: defaults.model,
     apiKey: defaults.apiKey ?? '',
+    apiFormat: 'chat_completions',
+    authHeader: '',
+    authHeaderValue: '',
   }
 }
 
@@ -174,7 +211,15 @@ function profileSummary(profile: ProviderProfile, isActive: boolean): string {
     models.length <= 3
       ? models.join(', ')
       : `${models[0]}, ${models[1]} + ${models.length - 2} more`
-  return `${providerKind} · ${profile.baseUrl} · ${modelDisplay} · ${keyInfo}${activeSuffix}`
+  const modeInfo =
+    profile.provider === 'openai'
+      ? ` · ${profile.apiFormat === 'responses' ? 'responses' : 'chat/completions'}`
+      : ''
+  const authInfo =
+    profile.provider === 'openai' && profile.authHeader
+      ? ` · ${profile.authHeader} auth`
+      : ''
+  return `${providerKind} · ${profile.baseUrl} · ${modelDisplay}${modeInfo}${authInfo} · ${keyInfo}${activeSuffix}`
 }
 
 function getGithubCredentialSourceFromEnv(
@@ -453,7 +498,18 @@ export function ProviderManager({ mode, onDone }: Props): React.ReactNode {
     })
   }, [])
 
-  const currentStep = FORM_STEPS[formStepIndex] ?? FORM_STEPS[0]
+  const formSteps = React.useMemo(
+    () =>
+      draftProvider === 'openai'
+        ? FORM_STEPS
+        : FORM_STEPS.filter(step =>
+            step.key !== 'apiFormat' &&
+            step.key !== 'authHeader' &&
+            step.key !== 'authHeaderValue'
+          ),
+    [draftProvider],
+  )
+  const currentStep = formSteps[formStepIndex] ?? formSteps[0] ?? FORM_STEPS[0]
   const currentStepKey = currentStep.key
   const currentValue = draft[currentStepKey]
 
@@ -671,17 +727,7 @@ export function ProviderManager({ mode, onDone }: Props): React.ReactNode {
   }
 
   function clearStartupProviderOverrideFromUserSettings(): string | null {
-    const { error } = updateSettingsForSource('userSettings', {
-      env: {
-        CLAUDE_CODE_USE_OPENAI: undefined as any,
-        CLAUDE_CODE_USE_GEMINI: undefined as any,
-        CLAUDE_CODE_USE_GITHUB: undefined as any,
-        CLAUDE_CODE_USE_BEDROCK: undefined as any,
-        CLAUDE_CODE_USE_VERTEX: undefined as any,
-        CLAUDE_CODE_USE_FOUNDRY: undefined as any,
-      },
-    })
-    return error ? error.message : null
+    return clearStartupProviderOverrides()
   }
 
   function buildCodexOAuthActivationMessage(options: {
@@ -768,12 +814,14 @@ export function ProviderManager({ mode, onDone }: Props): React.ReactNode {
           mainLoopModelForSession: null,
         }))
         refreshProfiles()
-        setAppState(prev => ({
-          ...prev,
-          mainLoopModel: GITHUB_PROVIDER_DEFAULT_MODEL,
-        }))
         setStatusMessage(`Active provider: ${GITHUB_PROVIDER_LABEL}`)
         setIsActivating(false)
+        onDone({
+          action: 'activated',
+          activeProviderName: GITHUB_PROVIDER_LABEL,
+          activeProviderModel: GITHUB_PROVIDER_DEFAULT_MODEL,
+          message: `Provider switched to ${GITHUB_PROVIDER_LABEL} (${GITHUB_PROVIDER_DEFAULT_MODEL})`,
+        })
         returnToMenu()
         return
       }
@@ -789,19 +837,14 @@ export function ProviderManager({ mode, onDone }: Props): React.ReactNode {
       // Update the session model to the new provider's first model.
       // persistActiveProviderProfileModel (called by onChangeAppState) will
       // not overwrite the multi-model list because it checks if the model
-      // is already in the profile's comma-separated model list.
+      // is already in the provider's configured model list.
       const newModel = getPrimaryModel(active.model)
       setAppState(prev => ({
         ...prev,
         mainLoopModel: newModel,
-      }))
-
-      providerLabel = active.name
-      setAppState(prev => ({
-        ...prev,
-        mainLoopModel: active.model,
         mainLoopModelForSession: null,
       }))
+      providerLabel = active.name
       const settingsOverrideError =
         clearStartupProviderOverrideFromUserSettings()
       const isActiveCodexOAuth = isCodexOAuthProfile(
@@ -813,23 +856,29 @@ export function ProviderManager({ mode, onDone }: Props): React.ReactNode {
         : null
 
       refreshProfiles()
-      setStatusMessage(
-        isActiveCodexOAuth
-          ? buildCodexOAuthActivationMessage({
-              prefix: `Active provider: ${active.name}`,
+      const activationMessage = isActiveCodexOAuth
+        ? buildCodexOAuthActivationMessage({
+            prefix: `Active provider: ${active.name}`,
+            activationWarning,
+            warnings: [
               activationWarning,
-              warnings: [
-                activationWarning,
-                settingsOverrideError
-                  ? `could not clear startup provider override (${settingsOverrideError})`
-                  : null,
-              ].filter((warning): warning is string => Boolean(warning)),
-            })
-          : settingsOverrideError
-            ? `Active provider: ${active.name}. Warning: could not clear startup provider override (${settingsOverrideError}).`
-            : `Active provider: ${active.name}`,
-      )
+              settingsOverrideError
+                ? `could not clear startup provider override (${settingsOverrideError})`
+                : null,
+            ].filter((warning): warning is string => Boolean(warning)),
+          })
+        : settingsOverrideError
+          ? `Active provider: ${active.name}. Warning: could not clear startup provider override (${settingsOverrideError}).`
+          : `Active provider: ${active.name}`
+      setStatusMessage(activationMessage)
       setIsActivating(false)
+      onDone({
+        action: 'activated',
+        activeProfileId: active.id,
+        activeProviderName: active.name,
+        activeProviderModel: newModel,
+        message: `Provider switched to ${active.name} (${newModel})`,
+      })
       returnToMenu()
     } catch (error) {
       refreshProfiles()
@@ -944,6 +993,9 @@ export function ProviderManager({ mode, onDone }: Props): React.ReactNode {
       baseUrl: defaults.baseUrl,
       model: defaults.model,
       apiKey: defaults.apiKey ?? '',
+      apiFormat: 'chat_completions',
+      authHeader: '',
+      authHeaderValue: '',
     }
     setEditingProfileId(null)
     setDraftProvider(defaults.provider ?? 'openai')
@@ -990,6 +1042,22 @@ export function ProviderManager({ mode, onDone }: Props): React.ReactNode {
       baseUrl: nextDraft.baseUrl,
       model: nextDraft.model,
       apiKey: nextDraft.apiKey,
+      apiFormat:
+        draftProvider === 'openai' && nextDraft.apiFormat === 'responses'
+          ? 'responses'
+          : 'chat_completions',
+      authHeader:
+        draftProvider === 'openai' && nextDraft.authHeader
+          ? nextDraft.authHeader
+          : undefined,
+      authScheme:
+        draftProvider === 'openai' && nextDraft.authHeader
+          ? (nextDraft.authHeader.toLowerCase() === 'authorization' ? 'bearer' : 'raw')
+          : undefined,
+      authHeaderValue:
+        draftProvider === 'openai' && nextDraft.authHeaderValue
+          ? nextDraft.authHeaderValue
+          : undefined,
     }
 
     const saved = editingProfileId
@@ -1005,7 +1073,7 @@ export function ProviderManager({ mode, onDone }: Props): React.ReactNode {
     if (isActiveSavedProfile) {
       setAppState(prev => ({
         ...prev,
-        mainLoopModel: saved.model,
+        mainLoopModel: getPrimaryModel(saved.model),
         mainLoopModelForSession: null,
       }))
     }
@@ -1212,9 +1280,9 @@ export function ProviderManager({ mode, onDone }: Props): React.ReactNode {
     setDraft(nextDraft)
     setErrorMessage(undefined)
 
-    if (formStepIndex < FORM_STEPS.length - 1) {
+    if (formStepIndex < formSteps.length - 1) {
       const nextIndex = formStepIndex + 1
-      const nextKey = FORM_STEPS[nextIndex]?.key ?? 'name'
+      const nextKey = formSteps[nextIndex]?.key ?? 'name'
       setFormStepIndex(nextIndex)
       setCursorOffset(nextDraft[nextKey].length)
       return
@@ -1228,7 +1296,7 @@ export function ProviderManager({ mode, onDone }: Props): React.ReactNode {
 
     if (formStepIndex > 0) {
       const nextIndex = formStepIndex - 1
-      const nextKey = FORM_STEPS[nextIndex]?.key ?? 'name'
+      const nextKey = formSteps[nextIndex]?.key ?? 'name'
       setFormStepIndex(nextIndex)
       setCursorOffset(draft[nextKey].length)
       return
@@ -1279,6 +1347,11 @@ export function ProviderManager({ mode, onDone }: Props): React.ReactNode {
         label: 'Azure OpenAI',
         description: 'Azure OpenAI endpoint (model=deployment name)',
       },
+      {
+        value: 'bankr',
+        label: 'Bankr',
+        description: 'Bankr LLM Gateway (OpenAI-compatible)',
+      },
       ...(canUseCodexOAuth
         ? [
             {
@@ -1321,8 +1394,13 @@ export function ProviderManager({ mode, onDone }: Props): React.ReactNode {
       },
       {
         value: 'moonshotai',
-        label: 'Moonshot AI',
-        description: 'Kimi OpenAI-compatible endpoint',
+        label: 'Moonshot AI - API',
+        description: 'Moonshot AI - API endpoint',
+      },
+      {
+        value: 'kimi-code',
+        label: 'Moonshot AI - Kimi Code',
+        description: 'Moonshot AI - Kimi Code Subscription endpoint',
       },
       {
         value: 'nvidia-nim',
@@ -1349,6 +1427,16 @@ export function ProviderManager({ mode, onDone }: Props): React.ReactNode {
         label: 'Together AI',
         description: 'Together chat/completions endpoint',
       },
+      {
+        value: 'xai',
+        label: 'xAI',
+        description: 'xAI Grok OpenAI-compatible endpoint',
+      },
+      {
+        value: 'zai',
+        label: 'Z.AI - GLM Coding Plan',
+        description: 'Z.AI GLM coding subscription endpoint',
+      },
       {
         value: 'custom',
         label: 'Custom',
@@ -1413,28 +1501,59 @@ export function ProviderManager({ mode, onDone }: Props): React.ReactNode {
             : 'OpenAI-compatible API'}
         </Text>
         <Text dimColor>
-          Step {formStepIndex + 1} of {FORM_STEPS.length}: {currentStep.label}
+          Step {formStepIndex + 1} of {formSteps.length}: {currentStep.label}
         </Text>
-        <Box flexDirection="row" gap={1}>
-          <Text>{figures.pointer}</Text>
-          <TextInput
-            value={currentValue}
-            onChange={value =>
-              setDraft(prev => ({
-                ...prev,
-                [currentStepKey]: value,
-              }))
+        {currentStepKey === 'apiFormat' ? (
+          <Select
+            options={[
+              {
+                value: 'chat_completions',
+                label: 'Chat Completions',
+                description: 'Use /chat/completions for broad OpenAI-compatible support',
+              },
+              {
+                value: 'responses',
+                label: 'Responses',
+                description: 'Use /responses for providers that support the Responses API',
+              },
+            ]}
+            defaultValue={
+              currentValue === 'responses' ? 'responses' : 'chat_completions'
             }
-            onSubmit={handleFormSubmit}
-            focus={true}
-            showCursor={true}
-            placeholder={`${currentStep.placeholder}${figures.ellipsis}`}
-            mask={currentStepKey === 'apiKey' ? '*' : undefined}
-            columns={80}
-            cursorOffset={cursorOffset}
-            onChangeCursorOffset={setCursorOffset}
+            defaultFocusValue={
+              currentValue === 'responses' ? 'responses' : 'chat_completions'
+            }
+            onChange={value => handleFormSubmit(value)}
+            onCancel={handleBackFromForm}
+            visibleOptionCount={2}
           />
-        </Box>
+        ) : (
+          <Box flexDirection="row" gap={1}>
+            <Text>{figures.pointer}</Text>
+            <TextInput
+              value={currentValue}
+              onChange={value =>
+                setDraft(prev => ({
+                  ...prev,
+                  [currentStepKey]: value,
+                }))
+              }
+              onSubmit={handleFormSubmit}
+              focus={true}
+              showCursor={true}
+              placeholder={`${currentStep.placeholder}${figures.ellipsis}`}
+              mask={
+                currentStepKey === 'apiKey' ||
+                currentStepKey === 'authHeaderValue'
+                  ? '*'
+                  : undefined
+              }
+              columns={80}
+              cursorOffset={cursorOffset}
+              onChangeCursorOffset={setCursorOffset}
+            />
+          </Box>
+        )}
         {errorMessage && <Text color="error">{errorMessage}</Text>}
         <Text dimColor>
           Press Enter to continue. Press Esc to go back.

src/components/ResumeTask.tsx

@@ -119,17 +119,17 @@ export function ResumeTask({
     return <Box flexDirection="column" padding={1}>
         <Box flexDirection="row">
           <Spinner />
-          <Text bold>Loading Claude Code sessions…</Text>
+          <Text bold>Loading OpenClaude sessions…</Text>
         </Box>
         <Text dimColor>
-          {retrying ? 'Retrying…' : 'Fetching your Claude Code sessions…'}
+          {retrying ? 'Retrying…' : 'Fetching your OpenClaude sessions…'}
         </Text>
       </Box>;
   }
   if (loadErrorType) {
     return <Box flexDirection="column" padding={1}>
         <Text bold color="error">
-          Error loading Claude Code sessions
+          Error loading OpenClaude sessions
         </Text>
 
         {renderErrorSpecificGuidance(loadErrorType)}
@@ -143,7 +143,7 @@ export function ResumeTask({
   if (sessions.length === 0) {
     return <Box flexDirection="column" padding={1}>
         <Text bold>
-          No Claude Code sessions found
+          No OpenClaude sessions found
           {currentRepo && <Text> for {currentRepo}</Text>}
         </Text>
         <Box marginTop={1}>
@@ -261,7 +261,7 @@ function renderErrorSpecificGuidance(errorType: LoadErrorType): React.ReactNode
         </Box>;
     case 'other':
       return <Box marginY={1} flexDirection="row">
-          <Text dimColor>Sorry, Claude Code encountered an error</Text>
+          <Text dimColor>Sorry, OpenClaude encountered an error</Text>
         </Box>;
   }
 }

src/components/Settings/Config.tsx

@@ -299,6 +299,26 @@ export function Config({
         enabled: toolHistoryCompressionEnabled
       });
     }
+  }, {
+    id: 'showCacheStats',
+    label: 'Cache stats display',
+    value: globalConfig.showCacheStats,
+    options: ['off', 'compact', 'full'],
+    type: 'enum' as const,
+    onChange(mode: string) {
+      const showCacheStats = (mode === 'off' || mode === 'compact' || mode === 'full' ? mode : 'compact') as 'off' | 'compact' | 'full';
+      saveGlobalConfig(current_cs => ({
+        ...current_cs,
+        showCacheStats
+      }));
+      setGlobalConfig({
+        ...getGlobalConfig(),
+        showCacheStats
+      });
+      logEvent('tengu_show_cache_stats_setting_changed', {
+        mode: showCacheStats as AnalyticsMetadata_I_VERIFIED_THIS_IS_NOT_CODE_OR_FILEPATHS
+      });
+    }
   }, {
     id: 'spinnerTipsEnabled',
     label: 'Show tips',

src/components/Settings/MiniMaxUsage.tsx

@@ -0,0 +1,249 @@
+import * as React from 'react'
+import { useEffect, useState } from 'react'
+
+import { useTerminalSize } from '../../hooks/useTerminalSize.js'
+import { Box, Text } from '../../ink.js'
+import { useKeybinding } from '../../keybindings/useKeybinding.js'
+import {
+  buildMiniMaxUsageRows,
+  fetchMiniMaxUsage,
+  type MiniMaxUsageData,
+  type MiniMaxUsageRow,
+} from '../../services/api/minimaxUsage.js'
+import { logError } from '../../utils/log.js'
+import { ConfigurableShortcutHint } from '../ConfigurableShortcutHint.js'
+import { Byline } from '../design-system/Byline.js'
+import { ProgressBar } from '../design-system/ProgressBar.js'
+
+const RESET_COUNTDOWN_REFRESH_MS = 30_000
+const PROGRESS_BAR_WIDTH = 18
+
+type MiniMaxUsageLimitBarProps = {
+  label: string
+  usedPercent: number
+  resetsAt?: string
+  extraSubtext?: string
+  maxWidth: number
+  nowMs: number
+}
+
+function formatCountdownDuration(ms: number): string {
+  const totalMinutes = Math.max(1, Math.ceil(ms / 60_000))
+  const days = Math.floor(totalMinutes / 1_440)
+  const hours = Math.floor((totalMinutes % 1_440) / 60)
+  const minutes = totalMinutes % 60
+
+  if (days > 0) {
+    return hours > 0 ? `${days}d ${hours}h` : `${days}d`
+  }
+
+  if (hours > 0) {
+    return minutes > 0 ? `${hours}h ${minutes}m` : `${hours}h`
+  }
+
+  return `${minutes}m`
+}
+
+function formatResetCountdown(
+  resetsAt: string | undefined,
+  nowMs: number,
+): string | undefined {
+  if (!resetsAt) return undefined
+
+  const resetMs = Date.parse(resetsAt)
+  if (!Number.isFinite(resetMs)) return undefined
+
+  const remainingMs = resetMs - nowMs
+  if (remainingMs <= 0) {
+    return 'Resetting now'
+  }
+
+  return `Resets in ${formatCountdownDuration(remainingMs)}`
+}
+
+function MiniMaxUsageLimitBar({
+  label,
+  usedPercent,
+  resetsAt,
+  extraSubtext,
+  maxWidth,
+  nowMs,
+}: MiniMaxUsageLimitBarProps): React.ReactNode {
+  const normalizedUsedPercent = Math.max(0, Math.min(100, usedPercent))
+  const usedText = `${Math.floor(normalizedUsedPercent)}% used`
+  const resetText = formatResetCountdown(resetsAt, nowMs)
+  const details = [usedText, extraSubtext].filter(
+    (part): part is string => Boolean(part),
+  )
+
+  return (
+    <Box flexDirection="column">
+      <Text>
+        <Text bold>{label}</Text>
+        {resetText ? <Text dimColor> · {resetText}</Text> : null}
+      </Text>
+      <Box flexDirection="row" gap={1}>
+        <ProgressBar
+          ratio={normalizedUsedPercent / 100}
+          width={Math.min(PROGRESS_BAR_WIDTH, Math.max(1, maxWidth))}
+          fillColor="rate_limit_fill"
+          emptyColor="rate_limit_empty"
+        />
+        {details.length > 0 ? <Text dimColor>{details.join(' · ')}</Text> : null}
+      </Box>
+    </Box>
+  )
+}
+
+function MiniMaxUsageTextRow({
+  label,
+  value,
+}: Extract<MiniMaxUsageRow, { kind: 'text' }>): React.ReactNode {
+  if (!value) {
+    return <Text bold>{label}</Text>
+  }
+
+  return (
+    <Text>
+      <Text bold>{label}</Text>
+      <Text dimColor> · {value}</Text>
+    </Text>
+  )
+}
+
+export function MiniMaxUsage(): React.ReactNode {
+  const [usage, setUsage] = useState<MiniMaxUsageData | null>(null)
+  const [error, setError] = useState<string | null>(null)
+  const [isLoading, setIsLoading] = useState(true)
+  const [nowMs, setNowMs] = useState(() => Date.now())
+  const { columns } = useTerminalSize()
+  const availableWidth = columns - 2
+  const maxWidth = Math.min(availableWidth, 80)
+
+  const loadUsage = React.useCallback(async () => {
+    setIsLoading(true)
+    setError(null)
+
+    try {
+      setUsage(await fetchMiniMaxUsage())
+    } catch (err) {
+      logError(err as Error)
+      setError(
+        err instanceof Error ? err.message : 'Failed to load MiniMax usage',
+      )
+    } finally {
+      setIsLoading(false)
+    }
+  }, [])
+
+  useEffect(() => {
+    void loadUsage()
+  }, [loadUsage])
+
+  useEffect(() => {
+    const interval = setInterval(() => {
+      setNowMs(Date.now())
+    }, RESET_COUNTDOWN_REFRESH_MS)
+
+    return () => clearInterval(interval)
+  }, [])
+
+  useKeybinding(
+    'settings:retry',
+    () => {
+      void loadUsage()
+    },
+    {
+      context: 'Settings',
+      isActive: !!error && !isLoading,
+    },
+  )
+
+  if (error) {
+    return (
+      <Box flexDirection="column" gap={1}>
+        <Text color="error">Error: {error}</Text>
+        <Text dimColor>
+          <Byline>
+            <ConfigurableShortcutHint
+              action="settings:retry"
+              context="Settings"
+              fallback="r"
+              description="retry"
+            />
+            <ConfigurableShortcutHint
+              action="confirm:no"
+              context="Settings"
+              fallback="Esc"
+              description="cancel"
+            />
+          </Byline>
+        </Text>
+      </Box>
+    )
+  }
+
+  if (!usage) {
+    return (
+      <Box flexDirection="column" gap={1}>
+        <Text dimColor>Loading MiniMax usage data…</Text>
+        <Text dimColor>
+          <ConfigurableShortcutHint
+            action="confirm:no"
+            context="Settings"
+            fallback="Esc"
+            description="cancel"
+          />
+        </Text>
+      </Box>
+    )
+  }
+
+  const rows =
+    usage.availability === 'available'
+      ? buildMiniMaxUsageRows(usage.snapshots)
+      : []
+
+  return (
+    <Box flexDirection="column" gap={1} width="100%">
+      {usage.planType ? <Text dimColor>Plan: {usage.planType}</Text> : null}
+
+      {usage.availability === 'unknown' ? (
+        <Text dimColor>{usage.message}</Text>
+      ) : rows.length === 0 ? (
+        <Text dimColor>
+          No MiniMax usage windows were returned for this account.
+        </Text>
+      ) : null}
+
+      {rows.map((row, index) =>
+        row.kind === 'window' ? (
+          <MiniMaxUsageLimitBar
+            key={`${row.label}-${index}`}
+            label={row.label}
+            usedPercent={row.usedPercent}
+            resetsAt={row.resetsAt}
+            extraSubtext={row.extraSubtext}
+            maxWidth={maxWidth}
+            nowMs={nowMs}
+          />
+        ) : (
+          <MiniMaxUsageTextRow
+            key={`${row.label}-${index}`}
+            label={row.label}
+            value={row.value}
+          />
+        ),
+      )}
+
+      <Text dimColor>
+        <ConfigurableShortcutHint
+          action="confirm:no"
+          context="Settings"
+          fallback="Esc"
+          description="cancel"
+        />
+      </Text>
+    </Box>
+  )
+}

src/components/Settings/UnsupportedUsage.tsx

@@ -0,0 +1,28 @@
+import * as React from 'react'
+
+import { Box, Text } from '../../ink.js'
+import { ConfigurableShortcutHint } from '../ConfigurableShortcutHint.js'
+
+type UnsupportedUsageProps = {
+  providerLabel: string
+}
+
+export function UnsupportedUsage({
+  providerLabel,
+}: UnsupportedUsageProps): React.ReactNode {
+  return (
+    <Box flexDirection="column" gap={1}>
+      <Text dimColor>
+        Usage details are not currently available for {providerLabel}.
+      </Text>
+      <Text dimColor>
+        <ConfigurableShortcutHint
+          action="confirm:no"
+          context="Settings"
+          fallback="Esc"
+          description="cancel"
+        />
+      </Text>
+    </Box>
+  )
+}

src/components/Settings/Usage.tsx

@@ -17,6 +17,8 @@ import { Byline } from '../design-system/Byline.js';
 import { ProgressBar } from '../design-system/ProgressBar.js';
 import { isEligibleForOverageCreditGrant, OverageCreditUpsell } from '../LogoV2/OverageCreditUpsell.js';
 import { CodexUsage } from './CodexUsage.js';
+import { MiniMaxUsage } from './MiniMaxUsage.js';
+import { UnsupportedUsage } from './UnsupportedUsage.js';
 type LimitBarProps = {
   title: string;
   limit: RateLimit;
@@ -266,9 +268,26 @@ function AnthropicUsage(): React.ReactNode {
     </Box>;
 }
 export function Usage(): React.ReactNode {
-  if (getAPIProvider() === 'codex') {
+  const provider = getAPIProvider();
+  if (provider === 'codex') {
     return <CodexUsage />;
   }
+  if (provider === 'minimax') {
+    return <MiniMaxUsage />;
+  }
+  if (provider !== 'firstParty') {
+    const providerLabel = {
+      openai: 'this OpenAI-compatible provider',
+      gemini: 'Google Gemini',
+      github: 'GitHub Models',
+      mistral: 'Mistral',
+      'nvidia-nim': 'NVIDIA NIM',
+      bedrock: 'AWS Bedrock',
+      vertex: 'Google Vertex AI',
+      foundry: 'Microsoft Foundry'
+    }[provider] ?? 'this provider';
+    return <UnsupportedUsage providerLabel={providerLabel} />;
+  }
   return <AnthropicUsage />;
 }
 type ExtraUsageSectionProps = {

src/components/StartupScreen.test.ts

@@ -0,0 +1,257 @@
+import { afterEach, beforeEach, describe, expect, test } from 'bun:test'
+import { detectProvider } from './StartupScreen.js'
+
+const ENV_KEYS = [
+  'CLAUDE_CODE_USE_OPENAI',
+  'CLAUDE_CODE_USE_GEMINI',
+  'CLAUDE_CODE_USE_GITHUB',
+  'CLAUDE_CODE_USE_BEDROCK',
+  'CLAUDE_CODE_USE_VERTEX',
+  'CLAUDE_CODE_USE_MISTRAL',
+  'OPENAI_BASE_URL',
+  'OPENAI_API_KEY',
+  'OPENAI_MODEL',
+  'GEMINI_MODEL',
+  'MISTRAL_MODEL',
+  'ANTHROPIC_MODEL',
+  'CLAUDE_MODEL',
+  'NVIDIA_NIM',
+  'MINIMAX_API_KEY',
+]
+
+const originalEnv: Record<string, string | undefined> = {}
+
+beforeEach(() => {
+  for (const key of ENV_KEYS) {
+    originalEnv[key] = process.env[key]
+    delete process.env[key]
+  }
+})
+
+afterEach(() => {
+  for (const key of ENV_KEYS) {
+    if (originalEnv[key] === undefined) {
+      delete process.env[key]
+    } else {
+      process.env[key] = originalEnv[key]
+    }
+  }
+})
+
+function setupOpenAIMode(baseUrl: string, model: string): void {
+  process.env.CLAUDE_CODE_USE_OPENAI = '1'
+  process.env.OPENAI_BASE_URL = baseUrl
+  process.env.OPENAI_MODEL = model
+  process.env.OPENAI_API_KEY = 'test-key'
+}
+
+// --- Issue #855: aggregator URL must win over vendor-prefixed model name ---
+
+describe('detectProvider — aggregator URL authoritative over model-name substring (#855)', () => {
+  test('OpenRouter + deepseek/deepseek-chat labels as OpenRouter', () => {
+    setupOpenAIMode('https://openrouter.ai/api/v1', 'deepseek/deepseek-chat')
+    expect(detectProvider().name).toBe('OpenRouter')
+  })
+
+  test('OpenRouter + moonshotai/kimi-k2 labels as OpenRouter', () => {
+    setupOpenAIMode('https://openrouter.ai/api/v1', 'moonshotai/kimi-k2')
+    expect(detectProvider().name).toBe('OpenRouter')
+  })
+
+  test('OpenRouter + mistralai/mistral-large labels as OpenRouter', () => {
+    setupOpenAIMode('https://openrouter.ai/api/v1', 'mistralai/mistral-large')
+    expect(detectProvider().name).toBe('OpenRouter')
+  })
+
+  test('OpenRouter + meta-llama/llama-3.3 labels as OpenRouter', () => {
+    setupOpenAIMode('https://openrouter.ai/api/v1', 'meta-llama/llama-3.3-70b-instruct')
+    expect(detectProvider().name).toBe('OpenRouter')
+  })
+
+  test('Together + deepseek-ai/DeepSeek-V3 labels as Together AI', () => {
+    setupOpenAIMode('https://api.together.xyz/v1', 'deepseek-ai/DeepSeek-V3')
+    expect(detectProvider().name).toBe('Together AI')
+  })
+
+  test('Together + meta-llama/Llama-3.3 labels as Together AI', () => {
+    setupOpenAIMode('https://api.together.xyz/v1', 'meta-llama/Llama-3.3-70B-Instruct-Turbo')
+    expect(detectProvider().name).toBe('Together AI')
+  })
+
+  test('Groq + deepseek-r1-distill-llama-70b labels as Groq', () => {
+    setupOpenAIMode('https://api.groq.com/openai/v1', 'deepseek-r1-distill-llama-70b')
+    expect(detectProvider().name).toBe('Groq')
+  })
+
+  test('Groq + llama-3.3-70b-versatile labels as Groq', () => {
+    setupOpenAIMode('https://api.groq.com/openai/v1', 'llama-3.3-70b-versatile')
+    expect(detectProvider().name).toBe('Groq')
+  })
+
+  test('Azure + any deepseek deployment labels as Azure OpenAI', () => {
+    setupOpenAIMode('https://my-resource.openai.azure.com/', 'deepseek-chat')
+    expect(detectProvider().name).toBe('Azure OpenAI')
+  })
+})
+
+// --- Direct vendor endpoints still label correctly (regression) ---
+
+describe('detectProvider — direct vendor endpoints', () => {
+  test('api.deepseek.com labels as DeepSeek', () => {
+    setupOpenAIMode('https://api.deepseek.com/v1', 'deepseek-chat')
+    expect(detectProvider().name).toBe('DeepSeek')
+  })
+
+  test('api.kimi.com labels as Moonshot AI - Kimi Code', () => {
+    setupOpenAIMode('https://api.kimi.com/coding/v1', 'kimi-for-coding')
+    expect(detectProvider().name).toBe('Moonshot AI - Kimi Code')
+  })
+
+  test('api.moonshot.cn labels as Moonshot AI - API', () => {
+    setupOpenAIMode('https://api.moonshot.cn/v1', 'moonshot-v1-8k')
+    expect(detectProvider().name).toBe('Moonshot AI - API')
+  })
+
+  test('api.mistral.ai labels as Mistral', () => {
+    setupOpenAIMode('https://api.mistral.ai/v1', 'mistral-large-latest')
+    expect(detectProvider().name).toBe('Mistral')
+  })
+
+  test('api.z.ai labels as Z.AI GLM', () => {
+    setupOpenAIMode('https://api.z.ai/api/coding/paas/v4', 'GLM-5.1')
+    expect(detectProvider().name).toBe('Z.AI - GLM')
+  })
+
+  test('default OpenAI URL + gpt-4o labels as OpenAI', () => {
+    setupOpenAIMode('https://api.openai.com/v1', 'gpt-4o')
+    expect(detectProvider().name).toBe('OpenAI')
+  })
+})
+
+// --- rawModel fallback for generic/custom endpoints ---
+
+describe('detectProvider — rawModel fallback when URL is generic', () => {
+  test('custom proxy + deepseek-chat falls back to DeepSeek', () => {
+    setupOpenAIMode('https://my-proxy.internal/v1', 'deepseek-chat')
+    expect(detectProvider().name).toBe('DeepSeek')
+  })
+
+  test('custom proxy + kimi-for-coding falls back to Moonshot AI - Kimi Code', () => {
+    setupOpenAIMode('https://my-proxy.internal/v1', 'kimi-for-coding')
+    expect(detectProvider().name).toBe('Moonshot AI - Kimi Code')
+  })
+
+  test('custom proxy + kimi-k2 falls back to Moonshot AI - API', () => {
+    setupOpenAIMode('https://my-proxy.internal/v1', 'kimi-k2-instruct')
+    expect(detectProvider().name).toBe('Moonshot AI - API')
+  })
+
+  test('custom proxy + llama-3.3 falls back to Meta Llama', () => {
+    setupOpenAIMode('https://my-proxy.internal/v1', 'llama-3.3-70b')
+    expect(detectProvider().name).toBe('Meta Llama')
+  })
+
+  test('custom proxy + mistral-large falls back to Mistral', () => {
+    setupOpenAIMode('https://my-proxy.internal/v1', 'mistral-large-latest')
+    expect(detectProvider().name).toBe('Mistral')
+  })
+
+  test('custom proxy + exact uppercase GLM ID falls back to Z.AI GLM', () => {
+    setupOpenAIMode('https://my-proxy.internal/v1', 'GLM-5.1')
+    expect(detectProvider().name).toBe('Z.AI - GLM')
+  })
+
+  test('custom proxy + lowercase glm ID stays generic OpenAI', () => {
+    setupOpenAIMode('https://my-proxy.internal/v1', 'glm-5.1')
+    expect(detectProvider().name).toBe('OpenAI')
+  })
+
+  test('DashScope lowercase glm ID is not mislabeled as Z.AI', () => {
+    setupOpenAIMode('https://dashscope.aliyuncs.com/compatible-mode/v1', 'glm-5.1')
+    expect(detectProvider().name).toBe('OpenAI')
+  })
+})
+
+// --- Explicit env flags win over URL heuristics ---
+
+describe('detectProvider — explicit dedicated-provider env flags', () => {
+  test('NVIDIA_NIM=1 overrides aggregator URL', () => {
+    setupOpenAIMode('https://openrouter.ai/api/v1', 'some-nim-model')
+    process.env.NVIDIA_NIM = '1'
+    expect(detectProvider().name).toBe('NVIDIA NIM')
+  })
+
+  test('MINIMAX_API_KEY overrides aggregator URL', () => {
+    setupOpenAIMode('https://openrouter.ai/api/v1', 'any-model')
+    process.env.MINIMAX_API_KEY = 'test-key'
+    expect(detectProvider().name).toBe('MiniMax')
+  })
+})
+
+// --- modelOverride from --model flag ---
+
+describe('detectProvider — modelOverride from --model flag', () => {
+  test('modelOverride overrides default Anthropic model', () => {
+    const result = detectProvider('claude-opus-4-6')
+    expect(result.name).toBe('Anthropic')
+    expect(result.model).toContain('opus')
+  })
+
+  test('modelOverride alias is resolved for Anthropic', () => {
+    const result = detectProvider('opus')
+    expect(result.name).toBe('Anthropic')
+    expect(result.model).toContain('opus')
+  })
+
+  test('modelOverride takes priority over ANTHROPIC_MODEL env var', () => {
+    process.env.ANTHROPIC_MODEL = 'claude-haiku-4-5-20251001'
+    const result = detectProvider('claude-opus-4-6')
+    expect(result.name).toBe('Anthropic')
+    expect(result.model).toContain('opus')
+  })
+
+  test('modelOverride takes priority over CLAUDE_MODEL env var', () => {
+    process.env.CLAUDE_MODEL = 'claude-haiku-4-5-20251001'
+    const result = detectProvider('claude-opus-4-6')
+    expect(result.name).toBe('Anthropic')
+    expect(result.model).toContain('opus')
+  })
+
+  test('modelOverride works for OpenAI provider', () => {
+    process.env.CLAUDE_CODE_USE_OPENAI = '1'
+    process.env.OPENAI_API_KEY = 'test-key'
+    process.env.OPENAI_MODEL = 'gpt-4o'
+    const result = detectProvider('gpt-4-turbo')
+    expect(result.model).toContain('gpt-4-turbo')
+  })
+
+  test('modelOverride works for Gemini provider', () => {
+    process.env.CLAUDE_CODE_USE_GEMINI = '1'
+    const result = detectProvider('gemini-2.5-pro')
+    expect(result.model).toBe('gemini-2.5-pro')
+  })
+
+  test('modelOverride works for Mistral provider', () => {
+    process.env.CLAUDE_CODE_USE_MISTRAL = '1'
+    const result = detectProvider('mistral-large-latest')
+    expect(result.model).toBe('mistral-large-latest')
+  })
+
+  test('modelOverride works for GitHub provider', () => {
+    process.env.CLAUDE_CODE_USE_GITHUB = '1'
+    const result = detectProvider('gpt-4o')
+    expect(result.model).toContain('gpt-4o')
+  })
+
+  test('undefined modelOverride preserves default behavior', () => {
+    const result = detectProvider(undefined)
+    expect(result.name).toBe('Anthropic')
+    expect(result.model).toContain('sonnet')
+  })
+
+  test('no argument preserves default behavior', () => {
+    const result = detectProvider()
+    expect(result.name).toBe('Anthropic')
+    expect(result.model).toContain('sonnet')
+  })
+})

src/components/StartupScreen.ts

@@ -9,6 +9,7 @@ import { isLocalProviderUrl, resolveProviderRequest } from '../services/api/prov
 import { getLocalOpenAICompatibleProviderLabel } from '../utils/providerDiscovery.js'
 import { getSettings_DEPRECATED } from '../utils/settings/settings.js'
 import { parseUserSpecifiedModel } from '../utils/model/model.js'
+import { containsExactZaiGlmModelId, isZaiBaseUrl } from '../utils/zaiProvider.js'
 
 declare const MACRO: { VERSION: string; DISPLAY_VERSION?: string }
 
@@ -83,33 +84,33 @@ const LOGO_CLAUDE = [
 
 // ─── Provider detection ───────────────────────────────────────────────────────
 
-function detectProvider(): { name: string; model: string; baseUrl: string; isLocal: boolean } {
+export function detectProvider(modelOverride?: string): { name: string; model: string; baseUrl: string; isLocal: boolean } {
   const useGemini = process.env.CLAUDE_CODE_USE_GEMINI === '1' || process.env.CLAUDE_CODE_USE_GEMINI === 'true'
   const useGithub = process.env.CLAUDE_CODE_USE_GITHUB === '1' || process.env.CLAUDE_CODE_USE_GITHUB === 'true'
   const useOpenAI = process.env.CLAUDE_CODE_USE_OPENAI === '1' || process.env.CLAUDE_CODE_USE_OPENAI === 'true'
   const useMistral = process.env.CLAUDE_CODE_USE_MISTRAL === '1' || process.env.CLAUDE_CODE_USE_MISTRAL === 'true'
 
   if (useGemini) {
-    const model = process.env.GEMINI_MODEL || 'gemini-2.0-flash'
+    const model = modelOverride || process.env.GEMINI_MODEL || 'gemini-2.0-flash'
     const baseUrl = process.env.GEMINI_BASE_URL || 'https://generativelanguage.googleapis.com/v1beta/openai'
     return { name: 'Google Gemini', model, baseUrl, isLocal: false }
   }
 
   if (useMistral) {
-    const model = process.env.MISTRAL_MODEL || 'devstral-latest'
+    const model = modelOverride || process.env.MISTRAL_MODEL || 'devstral-latest'
     const baseUrl = process.env.MISTRAL_BASE_URL || 'https://api.mistral.ai/v1'
     return { name: 'Mistral', model, baseUrl, isLocal: false }
   }
 
   if (useGithub) {
-    const model = process.env.OPENAI_MODEL || 'github:copilot'
+    const model = modelOverride || process.env.OPENAI_MODEL || 'github:copilot'
     const baseUrl =
       process.env.OPENAI_BASE_URL || 'https://api.githubcopilot.com'
     return { name: 'GitHub Copilot', model, baseUrl, isLocal: false }
   }
 
   if (useOpenAI) {
-    const rawModel = process.env.OPENAI_MODEL || 'gpt-4o'
+    const rawModel = modelOverride || process.env.OPENAI_MODEL || 'gpt-4o'
     const resolvedRequest = resolveProviderRequest({
       model: rawModel,
       baseUrl: process.env.OPENAI_BASE_URL,
@@ -117,30 +118,44 @@ function detectProvider(): { name: string; model: string; baseUrl: string; isLoc
     const baseUrl = resolvedRequest.baseUrl
     const isLocal = isLocalProviderUrl(baseUrl)
     let name = 'OpenAI'
-    if (/nvidia/i.test(baseUrl) || /nvidia/i.test(rawModel) || process.env.NVIDIA_NIM)
-      name = 'NVIDIA NIM'
-    else if (/minimax/i.test(baseUrl) || /minimax/i.test(rawModel) || process.env.MINIMAX_API_KEY)
-      name = 'MiniMax'
-    else if (resolvedRequest.transport === 'codex_responses' || baseUrl.includes('chatgpt.com/backend-api/codex'))
+    // Explicit dedicated-provider env flags win.
+    if (process.env.NVIDIA_NIM) name = 'NVIDIA NIM'
+    else if (process.env.MINIMAX_API_KEY) name = 'MiniMax'
+    else if (
+      resolvedRequest.transport === 'codex_responses' ||
+      baseUrl.includes('chatgpt.com/backend-api/codex')
+    )
       name = 'Codex'
-    else if (/moonshot/i.test(baseUrl) || /kimi/i.test(rawModel))
-      name = 'Moonshot (Kimi)'
-    else if (/deepseek/i.test(baseUrl) || /deepseek/i.test(rawModel))
-      name = 'DeepSeek'
-    else if (/openrouter/i.test(baseUrl))
-      name = 'OpenRouter'
-    else if (/together/i.test(baseUrl))
-      name = 'Together AI'
-    else if (/groq/i.test(baseUrl))
-      name = 'Groq'
-    else if (/mistral/i.test(baseUrl) || /mistral/i.test(rawModel))
-      name = 'Mistral'
-    else if (/azure/i.test(baseUrl))
-      name = 'Azure OpenAI'
-    else if (/llama/i.test(rawModel))
-      name = 'Meta Llama'
-    else if (isLocal)
-      name = getLocalOpenAICompatibleProviderLabel(baseUrl)
+    // Base URL is authoritative — must precede rawModel checks so aggregators
+    // (OpenRouter/Together/Groq) aren't mislabelled as DeepSeek/Kimi/etc.
+    // when routed to models whose IDs contain a vendor prefix. See issue #855.
+    else if (/openrouter/i.test(baseUrl)) name = 'OpenRouter'
+    else if (/together/i.test(baseUrl)) name = 'Together AI'
+    else if (/groq/i.test(baseUrl)) name = 'Groq'
+    else if (/azure/i.test(baseUrl)) name = 'Azure OpenAI'
+    else if (/nvidia/i.test(baseUrl)) name = 'NVIDIA NIM'
+    else if (/minimax/i.test(baseUrl)) name = 'MiniMax'
+    else if (/api\.kimi\.com/i.test(baseUrl)) name = 'Moonshot AI - Kimi Code'
+    else if (/moonshot/i.test(baseUrl)) name = 'Moonshot AI - API'
+    else if (/deepseek/i.test(baseUrl)) name = 'DeepSeek'
+    else if (/x\.ai/i.test(baseUrl)) name = 'xAI'
+    else if (isZaiBaseUrl(baseUrl)) name = 'Z.AI - GLM'
+    else if (/mistral/i.test(baseUrl)) name = 'Mistral'
+    // rawModel fallback — fires only when base URL is generic/custom.
+    else if (/nvidia/i.test(rawModel)) name = 'NVIDIA NIM'
+    else if (/minimax/i.test(rawModel)) name = 'MiniMax'
+    else if (/\bkimi-for-coding\b/i.test(rawModel))
+      name = 'Moonshot AI - Kimi Code'
+    else if (/\bkimi-k/i.test(rawModel) || /moonshot/i.test(rawModel))
+      name = 'Moonshot AI - API'
+    else if (/deepseek/i.test(rawModel)) name = 'DeepSeek'
+    else if (/grok/i.test(rawModel)) name = 'xAI'
+    else if (containsExactZaiGlmModelId(rawModel)) name = 'Z.AI - GLM'
+    else if (/mistral/i.test(rawModel)) name = 'Mistral'
+    else if (/llama/i.test(rawModel)) name = 'Meta Llama'
+    else if (/bankr/i.test(baseUrl)) name = 'Bankr'
+    else if (/bankr/i.test(rawModel)) name = 'Bankr'
+    else if (isLocal) name = getLocalOpenAICompatibleProviderLabel(baseUrl)
     
     // Resolve model alias to actual model name + reasoning effort
     let displayModel = resolvedRequest.resolvedModel
@@ -153,7 +168,7 @@ function detectProvider(): { name: string; model: string; baseUrl: string; isLoc
 
   // Default: Anthropic - check settings.model first, then env vars
   const settings = getSettings_DEPRECATED() || {}
-  const modelSetting = settings.model || process.env.ANTHROPIC_MODEL || process.env.CLAUDE_MODEL || 'claude-sonnet-4-6'
+  const modelSetting = modelOverride || settings.model || process.env.ANTHROPIC_MODEL || process.env.CLAUDE_MODEL || 'claude-sonnet-4-6'
   const resolvedModel = parseUserSpecifiedModel(modelSetting)
   const baseUrl = process.env.ANTHROPIC_BASE_URL ?? 'https://api.anthropic.com'
   const isLocal = isLocalProviderUrl(baseUrl)
@@ -169,11 +184,11 @@ function boxRow(content: string, width: number, rawLen: number): string {
 
 // ─── Main ─────────────────────────────────────────────────────────────────────
 
-export function printStartupScreen(): void {
+export function printStartupScreen(modelOverride?: string): void {
   // Skip in non-interactive / CI / print mode
   if (process.env.CI || !process.stdout.isTTY) return
 
-  const p = detectProvider()
+  const p = detectProvider(modelOverride)
   const W = 62
   const out: string[] = []
 

src/components/Stats.tsx

@@ -94,7 +94,7 @@ export function Stats(t0) {
   const allTimePromise = t1;
   let t2;
   if ($[1] === Symbol.for("react.memo_cache_sentinel")) {
-    t2 = <Box marginTop={1}><Spinner /><Text> Loading your Claude Code stats…</Text></Box>;
+    t2 = <Box marginTop={1}><Spinner /><Text> Loading your OpenClaude stats…</Text></Box>;
     $[1] = t2;
   } else {
     t2 = $[1];
@@ -242,7 +242,7 @@ function StatsContent(t0) {
   if (allTimeResult.type === "empty") {
     let t7;
     if ($[15] === Symbol.for("react.memo_cache_sentinel")) {
-      t7 = <Box marginTop={1}><Text color="warning">No stats available yet. Start using Claude Code!</Text></Box>;
+      t7 = <Box marginTop={1}><Text color="warning">No stats available yet. Start using OpenClaude!</Text></Box>;
       $[15] = t7;
     } else {
       t7 = $[15];

src/components/TeleportRepoMismatchDialog.tsx

@@ -73,7 +73,7 @@ export function TeleportRepoMismatchDialog(t0) {
   const options = t2;
   let t3;
   if ($[8] !== availablePaths.length || $[9] !== errorMessage || $[10] !== handleChange || $[11] !== options || $[12] !== targetRepo || $[13] !== validating) {
-    t3 = availablePaths.length > 0 ? <><Box flexDirection="column" gap={1}>{errorMessage && <Text color="error">{errorMessage}</Text>}<Text>Open Claude Code in <Text bold={true}>{targetRepo}</Text>:</Text></Box>{validating ? <Box><Spinner /><Text> Validating repository…</Text></Box> : <Select options={options} onChange={value_0 => void handleChange(value_0)} />}</> : <Box flexDirection="column" gap={1}>{errorMessage && <Text color="error">{errorMessage}</Text>}<Text dimColor={true}>Run claude --teleport from a checkout of {targetRepo}</Text></Box>;
+    t3 = availablePaths.length > 0 ? <><Box flexDirection="column" gap={1}>{errorMessage && <Text color="error">{errorMessage}</Text>}<Text>Open OpenClaude in <Text bold={true}>{targetRepo}</Text>:</Text></Box>{validating ? <Box><Spinner /><Text> Validating repository…</Text></Box> : <Select options={options} onChange={value_0 => void handleChange(value_0)} />}</> : <Box flexDirection="column" gap={1}>{errorMessage && <Text color="error">{errorMessage}</Text>}<Text dimColor={true}>Run openclaude --teleport from a checkout of {targetRepo}</Text></Box>;
     $[8] = availablePaths.length;
     $[9] = errorMessage;
     $[10] = handleChange;

src/components/TrustDialog/TrustDialog.tsx

@@ -206,7 +206,7 @@ export function TrustDialog(t0) {
   if ($[20] === Symbol.for("react.memo_cache_sentinel")) {
     t16 = <Text bold={true}>{getFsImplementation().cwd()}</Text>;
     t17 = <Text>Quick safety check: Is this a project you created or one you trust? (Like your own code, a well-known open source project, or work from your team). If not, take a moment to review what{"'"}s in this folder first.</Text>;
-    t18 = <Text>Claude Code{"'"}ll be able to read, edit, and execute files here.</Text>;
+    t18 = <Text>OpenClaude{"'"}ll be able to read, edit, and execute files here.</Text>;
     $[20] = t16;
     $[21] = t17;
     $[22] = t18;

src/components/mcp/ElicitationDialog.tsx

@@ -254,7 +254,7 @@ function ElicitationFormDialog({
   // Text fields are always in edit mode when focused — no Enter-to-edit step.
   const isEditingTextField = currentFieldIsText && !focusedButton;
   useRegisterOverlay('elicitation');
-  useNotifyAfterTimeout('Claude Code needs your input', 'elicitation_dialog');
+  useNotifyAfterTimeout('OpenClaude needs your input', 'elicitation_dialog');
 
   // Sync textInputValue when the focused field changes
   const syncTextInput = useCallback((fieldIndex: number | undefined) => {
@@ -1004,7 +1004,7 @@ function ElicitationURLDialog({
   const phaseRef = useRef<'prompt' | 'waiting'>('prompt');
   const [focusedButton, setFocusedButton] = useState<'accept' | 'decline' | 'open' | 'action' | 'cancel'>('accept');
   const showCancel = waitingState?.showCancel ?? false;
-  useNotifyAfterTimeout('Claude Code needs your input', 'elicitation_url_dialog');
+  useNotifyAfterTimeout('OpenClaude needs your input', 'elicitation_url_dialog');
   useRegisterOverlay('elicitation-url');
 
   // Keep refs in sync for use in abort handler (avoids re-registering listener)

src/components/mcp/MCPRemoteServerMenu.tsx

@@ -102,9 +102,9 @@ export function MCPRemoteServerMenu({
       if (success) {
         onComplete?.(`Authentication successful. Connected to ${server.name}.`);
       } else if (result.client.type === 'needs-auth') {
-        onComplete?.('Authentication successful, but server still requires authentication. You may need to manually restart Claude Code.');
+        onComplete?.('Authentication successful, but server still requires authentication. You may need to manually restart OpenClaude.');
       } else {
-        onComplete?.('Authentication successful, but server reconnection failed. You may need to manually restart Claude Code for the changes to take effect.');
+        onComplete?.('Authentication successful, but server reconnection failed. You may need to manually restart OpenClaude for the changes to take effect.');
       }
     } catch (err) {
       logEvent('tengu_claudeai_mcp_auth_completed', {
@@ -281,11 +281,11 @@ export function MCPRemoteServerMenu({
           const message = isEffectivelyAuthenticated ? `Authentication successful. Reconnected to ${server.name}.` : `Authentication successful. Connected to ${server.name}.`;
           onComplete?.(message);
         } else if (result_0.client.type === 'needs-auth') {
-          onComplete?.('Authentication successful, but server still requires authentication. You may need to manually restart Claude Code.');
+          onComplete?.('Authentication successful, but server still requires authentication. You may need to manually restart OpenClaude.');
         } else {
           // result.client.type === 'failed'
           logMCPDebug(server.name, `Reconnection failed after authentication`);
-          onComplete?.('Authentication successful, but server reconnection failed. You may need to manually restart Claude Code for the changes to take effect.');
+          onComplete?.('Authentication successful, but server reconnection failed. You may need to manually restart OpenClaude for the changes to take effect.');
         }
       }
     } catch (err_1) {

src/components/mcp/MCPSettings.tsx

@@ -147,7 +147,7 @@ export function MCPSettings(t0) {
         return;
       }
       if (servers.length === 0 && agentMcpServers.length === 0) {
-        onComplete("No MCP servers configured. Please run /doctor if this is unexpected. Otherwise, run `claude mcp --help` or visit https://code.claude.com/docs/en/mcp to learn more.");
+        onComplete("No MCP servers configured. Please run /doctor if this is unexpected. Otherwise, run `openclaude mcp --help` or visit https://github.com/Gitlawb/openclaude to learn more.");
       }
     };
     t8 = [servers.length, filteredClients.length, agentMcpServers.length, onComplete];

src/components/permissions/ComputerUseApproval/ComputerUseApproval.tsx

@@ -161,7 +161,7 @@ function ComputerUseTccPanel(t0) {
   }
   let t7;
   if ($[15] === Symbol.for("react.memo_cache_sentinel")) {
-    t7 = <Text dimColor={true}>Grant the missing permissions in System Settings, then select "Try again". macOS may require you to restart Claude Code after granting Screen Recording.</Text>;
+    t7 = <Text dimColor={true}>Grant the missing permissions in System Settings, then select "Try again". macOS may require you to restart OpenClaude after granting Screen Recording.</Text>;
     $[15] = t7;
   } else {
     t7 = $[15];

src/components/permissions/ExitPlanModePermissionRequest/ExitPlanModePermissionRequest.tsx

@@ -730,7 +730,7 @@ export function buildPlanApprovalOptions({
   });
   if (showUltraplan) {
     options.push({
-      label: 'No, refine with Ultraplan on Claude Code on the web',
+      label: 'No, refine with Ultraplan on OpenClaude on the web',
       value: 'ultraplan'
     });
   }

src/components/permissions/PermissionRequest.tsx

@@ -128,18 +128,18 @@ export type ToolUseConfirm<Input extends AnyObject = AnyObject> = {
 function getNotificationMessage(toolUseConfirm: ToolUseConfirm): string {
   const toolName = toolUseConfirm.tool.userFacingName(toolUseConfirm.input as never);
   if (toolUseConfirm.tool === ExitPlanModeV2Tool) {
-    return 'Claude Code needs your approval for the plan';
+    return 'OpenClaude needs your approval for the plan';
   }
   if (toolUseConfirm.tool === EnterPlanModeTool) {
-    return 'Claude Code wants to enter plan mode';
+    return 'OpenClaude wants to enter plan mode';
   }
   if (feature('REVIEW_ARTIFACT') && toolUseConfirm.tool === ReviewArtifactTool) {
-    return 'Claude needs your approval for a review artifact';
+    return 'OpenClaude needs your approval for a review artifact';
   }
   if (!toolName || toolName.trim() === '') {
-    return 'Claude Code needs your attention';
+    return 'OpenClaude needs your attention';
   }
-  return `Claude needs your permission to use ${toolName}`;
+  return `OpenClaude needs your permission to use ${toolName}`;
 }
 
 // TODO: Move this to Tool.renderPermissionRequest

src/components/permissions/rules/AddWorkspaceDirectory.tsx

@@ -40,7 +40,7 @@ function PermissionDescription() {
   const $ = _c(1);
   let t0;
   if ($[0] === Symbol.for("react.memo_cache_sentinel")) {
-    t0 = <Text dimColor={true}>Claude Code will be able to read files in this directory and make edits when auto-accept edits is on.</Text>;
+    t0 = <Text dimColor={true}>OpenClaude will be able to read files in this directory and make edits when auto-accept edits is on.</Text>;
     $[0] = t0;
   } else {
     t0 = $[0];

src/components/permissions/rules/PermissionRuleList.tsx

@@ -388,9 +388,9 @@ function PermissionRulesTab(t0) {
     let t8;
     if ($[10] === Symbol.for("react.memo_cache_sentinel")) {
       t8 = {
-        allow: "Claude Code won't ask before using allowed tools.",
-        ask: "Claude Code will always ask for confirmation before using these tools.",
-        deny: "Claude Code will always reject requests to use denied tools."
+        allow: "OpenClaude won't ask before using allowed tools.",
+        ask: "OpenClaude will always ask for confirmation before using these tools.",
+        deny: "OpenClaude will always reject requests to use denied tools."
       };
       $[10] = t8;
     } else {
@@ -1098,7 +1098,7 @@ export function PermissionRuleList(t0) {
   }
   let t28;
   if ($[89] === Symbol.for("react.memo_cache_sentinel")) {
-    t28 = <Text>Claude Code can read files in the workspace, and make edits when auto-accept edits is on.</Text>;
+    t28 = <Text>OpenClaude can read files in the workspace, and make edits when auto-accept edits is on.</Text>;
     $[89] = t28;
   } else {
     t28 = $[89];

src/components/permissions/rules/RemoveWorkspaceDirectory.tsx

@@ -68,7 +68,7 @@ export function RemoveWorkspaceDirectory(t0) {
   }
   let t4;
   if ($[10] === Symbol.for("react.memo_cache_sentinel")) {
-    t4 = <Text>Claude Code will no longer have access to files in this directory.</Text>;
+    t4 = <Text>OpenClaude will no longer have access to files in this directory.</Text>;
     $[10] = t4;
   } else {
     t4 = $[10];

src/components/tasks/RemoteSessionDetailDialog.tsx

@@ -44,7 +44,7 @@ type Props = {
 export function formatToolUseSummary(name: string, input: unknown): string {
   // plan_ready phase is only reached via ExitPlanMode tool
   if (name === EXIT_PLAN_MODE_V2_TOOL_NAME) {
-    return 'Review the plan in Claude Code on the web';
+    return 'Review the plan in OpenClaude on the web';
   }
   if (!input || typeof input !== 'object') return name;
   // AskUserQuestion: show the question text as a CTA, not the tool name.
@@ -168,7 +168,7 @@ function UltraplanSessionDetail(t0) {
     }
     let t7;
     if ($[12] === Symbol.for("react.memo_cache_sentinel")) {
-      t7 = <Text dimColor={true}>This will terminate the Claude Code on the web session.</Text>;
+      t7 = <Text dimColor={true}>This will terminate the OpenClaude on the web session.</Text>;
       $[12] = t7;
     } else {
       t7 = $[12];
@@ -311,7 +311,7 @@ function UltraplanSessionDetail(t0) {
   let t19;
   if ($[47] === Symbol.for("react.memo_cache_sentinel")) {
     t19 = {
-      label: "Review in Claude Code on the web",
+      label: "Review in OpenClaude on the web",
       value: "open" as const
     };
     $[47] = t19;
@@ -595,13 +595,13 @@ function ReviewSessionDetail(t0) {
   let t3;
   if ($[11] !== completed || $[12] !== onKill || $[13] !== running) {
     t3 = completed ? [{
-      label: "Open in Claude Code on the web",
+      label: "Open in OpenClaude on the web",
       value: "open"
     }, {
       label: "Dismiss",
       value: "dismiss"
     }] : [{
-      label: "Open in Claude Code on the web",
+      label: "Open in OpenClaude on the web",
       value: "open"
     }, ...(onKill && running ? [{
       label: "Stop ultrareview",

src/constants/promptIdentity.test.ts

@@ -11,6 +11,7 @@ import { afterEach, expect, test } from 'bun:test'
   NATIVE_PACKAGE_URL: undefined,
 }
 
+import { clearSystemPromptSections } from './systemPromptSections.js'
 import { getSystemPrompt, DEFAULT_AGENT_PROMPT } from './prompts.js'
 import { CLI_SYSPROMPT_PREFIXES, getCLISyspromptPrefix } from './system.js'
 import { CLAUDE_CODE_GUIDE_AGENT } from '../tools/AgentTool/built-in/claudeCodeGuideAgent.js'
@@ -23,6 +24,7 @@ const originalSimpleEnv = process.env.CLAUDE_CODE_SIMPLE
 
 afterEach(() => {
   process.env.CLAUDE_CODE_SIMPLE = originalSimpleEnv
+  clearSystemPromptSections()
 })
 
 test('CLI identity prefixes describe OpenClaude instead of Claude Code', () => {
@@ -47,6 +49,21 @@ test('simple mode identity describes OpenClaude instead of Claude Code', async (
   expect(prompt[0]).not.toContain("Anthropic's official CLI for Claude")
 })
 
+test('system prompt model identity updates when model changes mid-session', async () => {
+  delete process.env.CLAUDE_CODE_SIMPLE
+  clearSystemPromptSections()
+
+  const firstPrompt = await getSystemPrompt([], 'old-test-model')
+  const secondPrompt = await getSystemPrompt([], 'new-test-model')
+
+  const firstText = firstPrompt.join('\n')
+  const secondText = secondPrompt.join('\n')
+
+  expect(firstText).toContain('You are powered by the model old-test-model.')
+  expect(secondText).toContain('You are powered by the model new-test-model.')
+  expect(secondText).not.toContain('You are powered by the model old-test-model.')
+})
+
 test('built-in agent prompts describe OpenClaude instead of Claude Code', () => {
   expect(DEFAULT_AGENT_PROMPT).toContain('OpenClaude')
   expect(DEFAULT_AGENT_PROMPT).not.toContain('Claude Code')

src/constants/prompts.ts

@@ -496,7 +496,7 @@ ${CYBER_RISK_INSTRUCTION}`,
     systemPromptSection('ant_model_override', () =>
       getAntModelOverrideSection(),
     ),
-    systemPromptSection('env_info_simple', () =>
+    systemPromptSection(`env_info_simple:${model}`, () =>
       computeSimpleEnvInfo(model, additionalWorkingDirectories),
     ),
     systemPromptSection('language', () =>
@@ -519,7 +519,7 @@ ${CYBER_RISK_INSTRUCTION}`,
       'MCP servers connect/disconnect between turns',
     ),
     systemPromptSection('scratchpad', () => getScratchpadInstructions()),
-    systemPromptSection('frc', () => getFunctionResultClearingSection(model)),
+    systemPromptSection(`frc:${model}`, () => getFunctionResultClearingSection(model)),
     systemPromptSection(
       'summarize_tool_results',
       () => SUMMARIZE_TOOL_RESULTS_SECTION,

src/cost-tracker.cacheIntegration.test.ts

@@ -0,0 +1,128 @@
+/**
+ * Integration test for cost-tracker → cacheStatsTracker wiring.
+ *
+ * The unit tests in services/api/cacheMetrics.test.ts and
+ * services/api/cacheStatsTracker.test.ts verify that each piece works
+ * in isolation. This file verifies that they're ACTUALLY CONNECTED —
+ * that `addToTotalSessionCost` resolves the provider, extracts metrics,
+ * and records them on the tracker on every call. Without this test, a
+ * future refactor could silently unwire the call chain (wrong param
+ * order, renamed symbol, removed call) and every individual unit test
+ * would still pass while `/cache-stats` showed empty data.
+ *
+ * We use real state — `resetCostState` + `getCurrentTurnCacheMetrics` —
+ * rather than mocking the tracker module. Fewer moving parts, and the
+ * test fails for the right reason if anyone breaks the wrapping.
+ */
+import { beforeEach, describe, expect, test } from 'bun:test'
+import { addToTotalSessionCost, resetCostState } from './cost-tracker.js'
+import {
+  getCurrentTurnCacheMetrics,
+  getSessionCacheMetrics,
+} from './services/api/cacheStatsTracker.js'
+
+// BetaUsage-compatible shape — minimum fields addToTotalSessionCost
+// needs to run without throwing. Cache fields are the ones we care
+// about here; input/output go into model cost calc.
+function anthropicUsage(partial: {
+  input?: number
+  output?: number
+  cacheRead?: number
+  cacheCreation?: number
+}): Parameters<typeof addToTotalSessionCost>[1] {
+  return {
+    input_tokens: partial.input ?? 0,
+    output_tokens: partial.output ?? 0,
+    cache_read_input_tokens: partial.cacheRead ?? 0,
+    cache_creation_input_tokens: partial.cacheCreation ?? 0,
+    // BetaUsage has several other optional fields; they're not read by
+    // the cache-tracking path so we leave them undefined.
+  } as Parameters<typeof addToTotalSessionCost>[1]
+}
+
+beforeEach(() => {
+  // resetCostState is the wrapped version that ALSO clears the cache
+  // tracker — this line is itself part of what we're verifying.
+  resetCostState()
+})
+
+describe('addToTotalSessionCost → cacheStatsTracker wiring', () => {
+  test('records normalized cache metrics on the tracker for each call', () => {
+    addToTotalSessionCost(
+      0.01,
+      anthropicUsage({
+        input: 200,
+        output: 50,
+        cacheRead: 800,
+        cacheCreation: 100,
+      }),
+      'claude-sonnet-4',
+    )
+
+    const turn = getCurrentTurnCacheMetrics()
+    expect(turn.supported).toBe(true)
+    expect(turn.read).toBe(800)
+    expect(turn.created).toBe(100)
+    // total = fresh(200) + read(800) + created(100) = 1100
+    expect(turn.total).toBe(1_100)
+    // hitRate = read / total = 800 / 1100 ≈ 0.727
+    expect(turn.hitRate).toBeCloseTo(800 / 1_100, 4)
+  })
+
+  test('session aggregate accumulates across multiple API calls', () => {
+    addToTotalSessionCost(
+      0.01,
+      anthropicUsage({ input: 100, cacheRead: 400 }),
+      'claude-sonnet-4',
+    )
+    addToTotalSessionCost(
+      0.02,
+      anthropicUsage({ input: 200, cacheRead: 600 }),
+      'claude-sonnet-4',
+    )
+
+    const session = getSessionCacheMetrics()
+    expect(session.read).toBe(1_000)
+    // total = (100+400) + (200+600) = 1300
+    expect(session.total).toBe(1_300)
+    expect(session.hitRate).toBeCloseTo(1_000 / 1_300, 4)
+  })
+
+  test('cold turn (no cache read/created) still records as supported', () => {
+    addToTotalSessionCost(
+      0.005,
+      anthropicUsage({ input: 500, output: 100 }),
+      'claude-sonnet-4',
+    )
+
+    const turn = getCurrentTurnCacheMetrics()
+    expect(turn.supported).toBe(true)
+    expect(turn.read).toBe(0)
+    expect(turn.created).toBe(0)
+    expect(turn.total).toBe(500)
+    // hitRate computed against a non-zero total is 0, not null — empty
+    // cache on a cacheable provider is a legitimate "no-hit" signal.
+    expect(turn.hitRate).toBe(0)
+  })
+})
+
+describe('resetCostState wrapper also clears cache tracker', () => {
+  test('resetCostState() zeros both cost counters and cache stats', () => {
+    // Populate both systems
+    addToTotalSessionCost(
+      0.01,
+      anthropicUsage({ input: 100, cacheRead: 500 }),
+      'claude-sonnet-4',
+    )
+    expect(getSessionCacheMetrics().read).toBe(500)
+
+    // resetCostState is the WRAPPED version — bootstrap's
+    // resetCostState cleared cost state historically but not cache
+    // stats. The wrapper in cost-tracker.ts adds the second call.
+    resetCostState()
+
+    const session = getSessionCacheMetrics()
+    expect(session.read).toBe(0)
+    expect(session.supported).toBe(false)
+  })
+})

src/cost-tracker.ts

@@ -1,5 +1,14 @@
 import type { BetaUsage as Usage } from '@anthropic-ai/sdk/resources/beta/messages/messages.mjs'
 import chalk from 'chalk'
+import {
+  extractCacheMetrics,
+  resolveCacheProvider,
+} from './services/api/cacheMetrics.js'
+import {
+  recordRequest as recordCacheRequest,
+  resetSessionCacheStats,
+} from './services/api/cacheStatsTracker.js'
+import { getAPIProvider, isGithubNativeAnthropicMode } from './utils/model/providers.js'
 import {
   addToTotalCostState,
   addToTotalLinesChanged,
@@ -22,7 +31,7 @@ import {
   getTotalWebSearchRequests,
   getUsageForModel,
   hasUnknownModelCost,
-  resetCostState,
+  resetCostState as baseResetCostState,
   resetStateForTests,
   setCostStateForRestore,
   setHasUnknownModelCost,
@@ -62,12 +71,22 @@ export {
   formatCost,
   hasUnknownModelCost,
   resetStateForTests,
-  resetCostState,
   setHasUnknownModelCost,
   getModelUsage,
   getUsageForModel,
 }
 
+/**
+ * Wraps bootstrap's resetCostState() so /clear, /compact and session
+ * switches zero the cache-stats tracker alongside the cost counters.
+ * Exported under the same name so existing callers pick up the cache
+ * reset without any call-site changes.
+ */
+export function resetCostState(): void {
+  baseResetCostState()
+  resetSessionCacheStats()
+}
+
 type StoredCostState = {
   totalCostUSD: number
   totalAPIDuration: number
@@ -251,6 +270,16 @@ function round(number: number, precision: number): number {
   return Math.round(number * precision) / precision
 }
 
+// Env-gated verbose token usage log. Treated as a boolean regardless of
+// value specifics — any truthy-ish string switches it on. `verbose` is the
+// documented keyword but we accept `1`/`true` for ergonomic parity with
+// other OPENCLAUDE_* flags.
+function shouldLogTokenUsageVerbose(): boolean {
+  const v = (process.env.OPENCLAUDE_LOG_TOKEN_USAGE ?? '').trim().toLowerCase()
+  if (!v) return false
+  return v !== '0' && v !== 'false' && v !== 'off'
+}
+
 function addToTotalModelUsage(
   cost: number,
   usage: Usage,
@@ -287,6 +316,43 @@ export function addToTotalSessionCost(
   const modelUsage = addToTotalModelUsage(cost, usage, model)
   addToTotalCostState(cost, modelUsage, model)
 
+  // Record normalized cache metrics for REPL display + /cache-stats.
+  // Resolved from the current process provider — at this point `usage` has
+  // already been Anthropic-shaped by the shim layer, so we feed the
+  // corresponding bucket (anthropic / copilot-claude / openai-like) to the
+  // extractor. For providers that genuinely don't report cache data
+  // (vanilla Copilot, Ollama), resolveCacheProvider steers us to
+  // supported:false so the UI shows "N/A" instead of lying with "0%".
+  const cacheProvider = resolveCacheProvider(getAPIProvider(), {
+    githubNativeAnthropic: isGithubNativeAnthropicMode(model),
+    openAiBaseUrl: process.env.OPENAI_BASE_URL ?? process.env.OPENAI_API_BASE,
+  })
+  const cacheMetrics = extractCacheMetrics(
+    usage as unknown as Record<string, unknown>,
+    cacheProvider,
+  )
+  recordCacheRequest(cacheMetrics, model)
+
+  // Opt-in structured per-request debug log on stderr. Power-user knob, not
+  // shown in the REPL — complements CLAUDE_CODE_ENABLE_TOKEN_USAGE_ATTACHMENT
+  // (which is model-facing). Any truthy value except "0"/"false" enables it.
+  if (shouldLogTokenUsageVerbose()) {
+    process.stderr.write(
+      JSON.stringify({
+        tag: 'openclaude.tokenUsage',
+        model,
+        provider: cacheProvider,
+        input_tokens: usage.input_tokens,
+        output_tokens: usage.output_tokens,
+        cache_read_input_tokens: usage.cache_read_input_tokens ?? 0,
+        cache_creation_input_tokens: usage.cache_creation_input_tokens ?? 0,
+        cache_supported: cacheMetrics.supported,
+        cache_hit_rate: cacheMetrics.hitRate,
+        cost_usd: cost,
+      }) + '\n',
+    )
+  }
+
   const attrs =
     isFastModeEnabled() && usage.speed === 'fast'
       ? { model, speed: 'fast' }

src/entrypoints/cli.tsx

@@ -80,7 +80,7 @@ async function main(): Promise<void> {
   if (args.length === 1 && (args[0] === '--version' || args[0] === '-v' || args[0] === '-V')) {
     // MACRO.VERSION is inlined at build time
     // biome-ignore lint/suspicious/noConsole:: intentional console output
-    console.log(`${MACRO.DISPLAY_VERSION ?? MACRO.VERSION} (Open Claude)`);
+    console.log(`${MACRO.DISPLAY_VERSION ?? MACRO.VERSION} (OpenClaude)`);
     return;
   }
 
@@ -134,9 +134,13 @@ async function main(): Promise<void> {
 
   await validateProviderEnvForStartupOrExit()
 
+  // Parse --model early so the startup screen can display the override
+  const { eagerParseCliFlag } = await import('../utils/cliArgs.js')
+  const earlyModelFlag = eagerParseCliFlag('--model')
+
   // Print the gradient startup screen before the Ink UI loads
   const { printStartupScreen } = await import('../components/StartupScreen.js')
-  printStartupScreen()
+  printStartupScreen(earlyModelFlag)
 
   // For all other paths, load the startup profiler
   const {

src/hooks/notifs/useNpmDeprecationNotification.tsx

@@ -2,7 +2,7 @@ import { isInBundledMode } from 'src/utils/bundledMode.js';
 import { getCurrentInstallationType } from 'src/utils/doctorDiagnostic.js';
 import { isEnvTruthy } from 'src/utils/envUtils.js';
 import { useStartupNotification } from './useStartupNotification.js';
-const NPM_DEPRECATION_MESSAGE = 'Claude Code has switched from npm to native installer. Run `claude install` or see https://docs.anthropic.com/en/docs/claude-code/getting-started for more options.';
+const NPM_DEPRECATION_MESSAGE = 'OpenClaude has switched from npm to the native installer. Run `openclaude install` or see https://github.com/Gitlawb/openclaude#quick-start for more options.';
 export function useNpmDeprecationNotification() {
   useStartupNotification(_temp);
 }

src/hooks/useDiffInIDE.ts

@@ -60,7 +60,7 @@ export function useDiffInIDE({
 
   const sha = useMemo(() => randomUUID().slice(0, 6), [])
   const tabName = useMemo(
-    () => `✻ [Claude Code] ${basename(filePath)} (${sha}) ⧉`,
+    () => `✻ [OpenClaude] ${basename(filePath)} (${sha}) ⧉`,
     [filePath, sha],
   )
 

src/hooks/usePasteHandler.test.ts

@@ -1,5 +1,8 @@
 import { expect, test } from 'bun:test'
-import { supportsClipboardImageFallback } from './usePasteHandler.ts'
+import {
+  shouldHandleInputAsPaste,
+  supportsClipboardImageFallback,
+} from './usePasteHandler.ts'
 
 test('supports clipboard image fallback on Windows', () => {
   expect(supportsClipboardImageFallback('windows')).toBe(true)
@@ -20,3 +23,42 @@ test('does not support clipboard image fallback on WSL', () => {
 test('does not support clipboard image fallback on unknown platforms', () => {
   expect(supportsClipboardImageFallback('unknown')).toBe(false)
 })
+
+test('does not treat a bracketed paste as pending when no paste handlers are provided', () => {
+  expect(
+    shouldHandleInputAsPaste({
+      hasTextPasteHandler: false,
+      hasImagePasteHandler: false,
+      inputLength: 'kimi-k2.5'.length,
+      pastePending: false,
+      hasImageFilePath: false,
+      isFromPaste: true,
+    }),
+  ).toBe(false)
+})
+
+test('treats bracketed text paste as pending when a text paste handler exists', () => {
+  expect(
+    shouldHandleInputAsPaste({
+      hasTextPasteHandler: true,
+      hasImagePasteHandler: false,
+      inputLength: 'kimi-k2.5'.length,
+      pastePending: false,
+      hasImageFilePath: false,
+      isFromPaste: true,
+    }),
+  ).toBe(true)
+})
+
+test('treats image path paste as pending when only an image handler exists', () => {
+  expect(
+    shouldHandleInputAsPaste({
+      hasTextPasteHandler: false,
+      hasImagePasteHandler: true,
+      inputLength: 'C:\\Users\\jat\\image.png'.length,
+      pastePending: false,
+      hasImageFilePath: true,
+      isFromPaste: false,
+    }),
+  ).toBe(true)
+})