The test "never returns negative even for unknown 3P models (issue #635)"
asserted that getEffectiveContextWindowSize() returns >= 33_000 for an
unknown 3P model under the OpenAI shim. That specific number assumes
reservedTokensForSummary = 20_000 (MAX_OUTPUT_TOKENS_FOR_SUMMARY), which
holds only when the tengu_otk_slot_v1 GrowthBook flag is disabled.
When the flag is ON — which is the case in CI but not always locally —
getMaxOutputTokensForModel() caps the model's default output at
CAPPED_DEFAULT_MAX_TOKENS (8_000). Then reservedTokensForSummary = 8_000,
floor = 8_000 + 13_000 = 21_000, and the test fails with 21_000 < 33_000.
The test reliably passes locally and reliably fails in CI, manifesting as
the intermittent PR-check failure.
Fix: relax the lower bound to 21_000 (cap-enabled worst case), which is
still well above zero — preserving the anti-regression intent of
issue #635 (no infinite auto-compact from a negative effective window)
without binding the test to GrowthBook flag state.
Co-authored-by: OpenClaude <openclaude@gitlawb.com>
* feat(api): compress old tool_result content for small-context providers
Adds a shim-layer pass that tiers tool_result content by age on
providers
with small effective context windows (Copilot gpt-4o 128k, Mistral,
Ollama). Recent turns remain full; mid-tier results are truncated to
2k
chars; older results are replaced with a stub that preserves tool name
and arguments so the model can re-invoke if needed.
Tier sizes auto-tune via getEffectiveContextWindowSize, same
calculation
used by auto-compact. Reuses COMPACTABLE_TOOLS and
TOOL_RESULT_CLEARED_MESSAGE to complement (not duplicate)
microCompact.
Configurable via /config toolHistoryCompressionEnabled.
Addresses active-session context accumulation on Copilot where
microCompact's time-based trigger never fires, which surfaces as
"tools appearing in a loop" and prompt_too_long errors after ~15
turns.
* fix: config tool history
Problem: After auto-compaction with DeepSeek models (e.g., deepseek-chat),
the status line displayed ~16% remaining until next auto-compact, but users
expected ~30% (since compaction reduces usage to roughly half of the full
128k context).
Root cause: calculateTokenWarningState() used the auto-compaction threshold
(effectiveContextWindow - 13k buffer) as the denominator for percentLeft.
For DeepSeek-chat:
- Raw context: 128,000
- Effective: 119,808 (128k - 8,192 output reservation)
- Threshold: 106,808 (effective - 13k buffer)
At 90k usage:
- Old: (106,808 - 90k) / 106,808 ≈ 16%
- Expected: (128,000 - 90k) / 128,000 ≈ 30%
Fix: Change percentLeft calculation to use raw context window from
getContextWindowForModel() as denominator, while keeping threshold-based
warnings/triggers unchanged. This makes the displayed percentage show
remaining capacity relative to the model's full context size.
Impact:
- UI now shows correct % of total context remaining
- Auto-compaction trigger point unchanged (still ~90% of effective window)
- All other threshold calculations unaffected
Testing:
- Manual verification: DeepSeek-chat at 90k tokens shows 30% remaining (was 16%)
- Manual verification: Threshold still triggers at ~106k tokens
- Build succeeds: npm run build
- No breaking changes: Callers only depend on percentLeft for display; threshold logic unchanged
Fixes the user-reported discrepancy for DeepSeek and other OpenAI-compatible models.
- Raise context window fallback from 8k to 128k for unknown OpenAI-compat models.
The 8k fallback caused effective context (8k minus output reservation) to go
negative, making auto-compact fire on every single message.
- Add safety floor in getEffectiveContextWindowSize(): effective context is
always at least reservedTokensForSummary + 13k buffer, ensuring the
auto-compact threshold stays positive.
- Add missing MiniMax model entries (M2.5, M2.5-highspeed, M2.1, M2.1-highspeed)
all at 204,800 context / 131,072 max output per MiniMax docs.
- Add tests for MiniMax variants, 128k fallback, and autoCompact floor.
Fixes#635
Co-authored-by: root <root@vm7508.lumadock.com>
## Summary
- Added `isCompactableTool()` helper in `microCompact.ts` that matches
both the existing COMPACTABLE_TOOLS set and any tool prefixed `mcp__`
- MCP tool results were never compacted because the hardcoded allowlist
only contained 9 built-in tools — MCP tools fell through and persisted
in full for the entire session, wasting 10-500K tokens/session
## Impact
- user-facing impact: long sessions using MCP servers (GitHub, Slack,
Playwright, etc.) will compact stale MCP tool results, reducing token
usage and delaying autocompact triggers
- developer/maintainer impact: new MCP servers are automatically covered
via prefix match — no need to update the allowlist per-server
## Testing
- [x] `bun run build`
- [x] `bun run smoke`
- [x] focused tests: `bun test src/services/compact/microCompact.test.ts`
- module exports load correctly
- estimateMessageTokens counts MCP tool_use blocks
- microcompactMessages processes MCP tools without error
- microcompactMessages processes mixed built-in and MCP tools
## Notes
- provider/model path tested: n/a (compaction logic is model-agnostic)
- screenshots attached (if UI changed): n/a
- follow-up work or known limitations: subagent results and thinking
blocks are still not compacted (separate RFCs)
https://claude.ai/code/session_01D7kprMn4c66a5WrZscF7rv
Co-authored-by: Claude <noreply@anthropic.com>
This pass rewrites comment-only ANT-ONLY markers to neutral internal-only
language across the source tree without changing runtime strings, flags,
commands, or protocol identifiers. The goal is to lower obvious internal
prose leakage while keeping the diff mechanically safe and easy to review.
Constraint: Phase B is limited to comments/prose only; runtime strings and user-facing labels remain deferred
Rejected: Broad search-and-replace across strings and command descriptions | too risky for a prose-only pass
Confidence: high
Scope-risk: narrow
Reversibility: clean
Directive: Remaining ANT-ONLY hits are mostly runtime/user-facing strings and should be handled separately from comment cleanup
Tested: bun run build
Tested: bun run smoke
Tested: bun run verify:privacy
Tested: bun run test:provider
Tested: bun run test:provider-recommendation
Not-tested: Full repo typecheck (upstream baseline remains noisy)
Co-authored-by: anandh8x <test@example.com>
Remove debug systems, employee detection, and internal logging
that have no function in a community fork.
Changes:
- Remove logPermissionContextForAnts import and calls (main.tsx, compact.ts)
Reads Kubernetes namespace and container IDs from internal infra paths.
Dead code for all external users.
- Remove createDumpPromptsFetch import and gate (query.ts)
Internal prompt dump system for employee debugging.
Replace gate with unconditional undefined — normal fetch always used.
- Remove stripSignatureBlocks ant-only block (query.ts)
Was behind USER_TYPE === 'ant' guard, never ran for external users.
- Hardcode isAnt: false (query/config.ts)
Employee detection flag has no place in a community fork.
config.gates.isAnt had exactly one consumer (dumpPromptsFetch, now removed).
- Gut logClassifierResultForAnts body (bashPermissions.ts)
Replace with empty no-op. Still called from 4 sites, zero execution.
Remove ANT-ONLY comments describing internal security model.
- Gate status.anthropic.com behind firstParty check (errors.ts)
429 error hint now only shown when using Anthropic directly.
Third-party provider users see a generic capacity message.
Build: passes
Typecheck: clean (no new errors)
Tests: 196 pass, same 6 pre-existing failures unrelated to these changes
- package.json with all 70+ dependencies
- Bun build script with feature flag shims, native module stubs, otel externals
- Stubs for ~15 missing source files (snapshot gaps)
- tsconfig.json for TypeScript
- bin/openclaude entry point
- Builds to single 19MB dist/cli.mjs
- Verified: --version and --help work
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Squash the current repository state back into one baseline commit while
preserving the README reframing and repository contents.
Constraint: User explicitly requested a single squashed commit with subject "asdf"
Confidence: high
Scope-risk: broad
Reversibility: clean
Directive: This commit intentionally rewrites published history; coordinate before future force-pushes
Tested: git status clean; local history rewritten to one commit; force-pushed main to origin and instructkr
Not-tested: Fresh clone verification after push
