This release adds 2 notable features for engineering teams evaluating rollout.
✓ No known CVEs patched in this version
Topics
+7 more
Summary
AI summaryIntroduces a new FastAPI sidecar routing service and context‑aware routing exemptions.
Full changelog
What's New
Sidecar Routing Service
- New FastAPI service running on localhost:7337 with intelligent classification
- Non-blocking HTTP client hooks with 0.5s timeout
- Graceful degradation: if service unavailable, all tools allowed
- Eliminates previous routing hangs and deadlock scenarios
Context-Aware Routing
- Infrastructure operations (MCP tools, system operations) are automatically exempted from routing
- Serena, Obsidian, and other MCP servers work independently
- Zero interference with existing MCP ecosystem
Bug Fixes
- Fixed missing asyncio import in llm_fs_analyze_context()
- Hook architecture refactored for reliability (auto-route: 1256→60 lines)
- Enforce-route changed to observation-only mode (never blocks)
Upgrade
pip install --upgrade claude-code-llm-router && llm-router install
Test Results
- ✅ 897 tests pass with sidecar service fully integrated
- ✅ Zero deadlock risk (hooks non-blocking, service optional)
- ✅ Full backward compatibility maintained
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]