Neutralize internal Anthropic prose in explanatory comments (#357)

This is a small prose-only follow-up that rewrites clearly internal or
explanatory Anthropic comment language to neutral wording in a handful of
high-confidence files. It avoids runtime strings, flags, command labels,
protocol identifiers, and provider-facing references.

Constraint: Keep this pass narrowly scoped to comments/documentation only
Rejected: Broader Anthropic comment sweep across functional API/protocol references | too ambiguous for a safe prose-only PR
Confidence: high
Scope-risk: narrow
Reversibility: clean
Directive: Leave functional Anthropic references (API behavior, SDKs, URLs, provider labels, protocol docs) for separate reviewed passes
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>

src/utils/attachments.ts

@@ -2641,7 +2641,7 @@ let suppressNext = false
 const FILTERED_LISTING_MAX = 30
 
 /**
- * Filter skills to bundled (Anthropic-curated) + MCP (user-connected) only.
+ * Filter skills to bundled + MCP (user-connected) only.
  * Used when skill-search is enabled to resolve the turn-0 gap for subagents:
  * these sources are small, intent-signaled, and won't hit the truncation budget.
  * User/project/plugin skills (the long tail — 200+) go through discovery instead.

src/utils/autoUpdater.ts

@@ -74,7 +74,7 @@ export async function assertMinVersion(): Promise<void> {
   }
 
   // Skip version check for third-party providers — the min version
-  // kill-switch is Anthropic-specific and should not block 3P users
+  // kill-switch is first-party-specific and should not block 3P users
   if (getAPIProvider() !== 'firstParty') {
     return
   }

src/utils/buildConfig.ts

@@ -1,7 +1,7 @@
 /**
  * OpenClaude build-time constants.
  *
- * These replace process.env checks that were only meaningful in Anthropic's
+ * These replace process.env checks that were only meaningful in the upstream
  * internal build. In OpenClaude all such gates are permanently disabled so
  * external users cannot activate internal code paths by setting env vars.
  */
@@ -9,7 +9,7 @@
 /**
  * Always false in OpenClaude.
  * Replaces all `process.env.USER_TYPE === 'ant'` checks so that no external
- * user can activate Anthropic-internal features (commit attribution hooks,
+ * user can activate internal-only features (commit attribution hooks,
  * system-prompt section clearing, dangerously-skip-permissions bypass, etc.)
  * by setting USER_TYPE in their shell environment.
  */

src/utils/http.ts

@@ -51,7 +51,7 @@ export function getMCPUserAgent(): string {
 }
 
 // User-Agent for WebFetch requests to arbitrary sites. `Claude-User` is
-// Anthropic's publicly documented agent for user-initiated fetches (what site
+// The first-party provider's publicly documented agent for user-initiated fetches (what site
 // operators match in robots.txt); the claude-code suffix lets them distinguish
 // local CLI traffic from claude.ai server-side fetches.
 export function getWebFetchUserAgent(): string {

src/utils/modifiers.ts

@@ -3,7 +3,7 @@ export type ModifierKey = 'shift' | 'command' | 'control' | 'option'
 /**
  * Pre-warm the native module by loading it in advance.
  *
- * NOTE: The `modifiers-napi` package is an Anthropic-internal native addon
+ * NOTE: The `modifiers-napi` package is an internal-only native addon
  * that is not shipped with the open-source build. All calls are no-ops here
  * to avoid supply-chain risk from unverified npm packages with the same name.
  */

src/utils/permissions/dangerousPatterns.ts

@@ -50,7 +50,7 @@ export const DANGEROUS_BASH_PATTERNS: readonly string[] = [
   'env',
   'xargs',
   'sudo',
-  // Anthropic internal: internal-only tools plus general tools that ant sandbox
+  // Internal-only: internal-only tools plus general tools that ant sandbox
   // dotfile data shows are commonly over-allowlisted as broad prefixes.
   // These stay internal-only — external users don't have coo, and the rest are
   // an empirical-risk call grounded in ant sandbox data, not a universal

src/utils/permissions/yoloClassifier.ts

@@ -98,7 +98,7 @@ export type AutoModeRules = {
  * captured tag contents ARE the defaults. Bullet items are single-line in the
  * template; each line starting with `- ` becomes one array entry.
  * Used by `claude auto-mode defaults`. Always returns external defaults,
- * never the Anthropic-internal template.
+ * never the internal-only template.
  */
 export function getDefaultExternalAutoModeRules(): AutoModeRules {
   return {