This release includes breaking changes for platform teams planning a safe upgrade.
✓ No known CVEs patched in this version
Topics
+7 more
Summary
AI summaryDeadlock eliminated when routing directive complexity mismatch occurs due to quota exhaustion.
Full changelog
What's new
- Pressure-Aware Complexity Downgrade — Automatically reduce task complexity when Claude subscription budget is exhausted (complex→moderate, moderate→simple at ≥95% quota)
- Stale Directive Remediation — Enforce-route hook now corrects outdated routing directives that were issued before quota exhaustion
Fixes
Eliminates deadlock scenario where:
- Routing directive issued when Sonnet at 85% quota
- By tool invocation time, Sonnet exhausted at 99%+
- Internal router tries Sonnet first, fails, forces fallback
- User stuck until manual
/model haikuoverride
With v5.8.0, directives issued with already-correct complexity from the start.
Upgrade
pip install --upgrade claude-code-llm-router && llm-router install
Technical Details
Auto-route hook (v21→v22): Pressure-aware downgrade applied before directive issuance
Enforce-route hook (v12→v13): Detects and corrects stale directives based on current budget
Test coverage: 900+ tests passing, all pressure downgrade scenarios validated
Backward compatible: No breaking changes to routing behavior
Weekly OSS security release digest.
The CVE patches and breaking changes that affected production tools this week. One email, every Sunday.
No spam, unsubscribe anytime.
Share this release
About ypollak2/llm-router
Subscription-aware LLM router for Claude Code. Routes tasks to 20+ providers (OpenAI, Gemini, Groq, Ollama, Codex) based on complexity classification, Claude subscription pressure, and cost. Free tasks stay on Claude subscription; expensive tasks fall back to the cheapest capable model. Includes 30 MCP tools, 6 auto-routing hooks, semantic dedup cache, prompt caching, daily spend cap, and a live web dashboard.
Related context
Related tools
Beta — feedback welcome: [email protected]