This release fixes issues for SREs watching stability and regressions.
✓ No known CVEs patched in this version
Topics
+7 more
Affected surfaces
ReleasePort's take
Light signalllm_doctor v8.5.1 fixes validation bugs and widens Python support to >=3.10, removing the <3.14 cap.
Why it matters: If your deployment uses llm_doctor with hooks or relies on Python versions between 3.10 and 3.14, upgrade to v8.5.1 immediately to eliminate silent failures and ensure compatibility.
Summary
AI summaryFixed silent failures in llm_doctor validation and widened Python version support to >=3.10.
Changes in this release
| Type | Severity | Summary | CVE |
|---|---|---|---|
| Bugfix | Medium |
llm_doctor validates hook Python interpreter paths exist llm_doctor validates hook Python interpreter paths exist Source: llm_adapter@2026-05-21 Confidence: high |
— |
| Bugfix | Medium |
llm_doctor detects duplicate hook entries in settings.json llm_doctor detects duplicate hook entries in settings.json Source: llm_adapter@2026-05-21 Confidence: high |
— |
| Bugfix | Medium |
Python version constraint widened to >=3.10, removes <3.14 cap Python version constraint widened to >=3.10, removes <3.14 cap Source: llm_adapter@2026-05-21 Confidence: low |
— |
Full changelog
Fixed
llm_doctornow validates hook Python interpreter paths exist — catches the #1 silent failure mode where routing breaks after moving a venvllm_doctordetects duplicate hook entries insettings.json- Python version constraint widened to
>=3.10(removes<3.14cap) — fixespipx installon Python 3.14
Upgrade
pip install --upgrade llm-routing
llm-router doctor # verify your installation is healthy
Full Changelog: https://github.com/ypollak2/llm-router/compare/v8.5.0...v8.5.1
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]