ypollak2/llm-router

v9.1.1 Bugfix

This release fixes issues for SREs watching stability and regressions.

Published 9d LLM Frameworks

View tool

✓ No known CVEs patched

Read the diff → Tool health → What is this tool? →

✓ No known CVEs patched in this version

Topics

ai-routing anthropic claude claude-code cost-optimization gemini

+7 more

litellm llm llm-router mcp-server model-router ollama openai

Summary

AI summary

Fixed live‑path routing for read‑only tool access, corrected strict‑mode banner display, improved hook installer compatibility, and resolved a quota pressure calculation error.

Changes in this release

Type	Severity	Summary	CVE
Bugfix
Bugfix	Medium	Read-only tool routing now uses external agent path for prompts like "Read notes.txt...". Read-only tool routing now uses external agent path for prompts like "Read notes.txt...". Source: llm_adapter@2026-05-25 Confidence: high	—
Bugfix	Medium	Strict-mode status banner correctly reports "strict zero-Claude routing" on session startup. Strict-mode status banner correctly reports "strict zero-Claude routing" on session startup. Source: llm_adapter@2026-05-25 Confidence: high	—
Bugfix	Medium	Hook installer recognizes Claude Code's nested hook settings schema and no longer crashes on reinstall. Hook installer recognizes Claude Code's nested hook settings schema and no longer crashes on reinstall. Source: llm_adapter@2026-05-25 Confidence: high	—
Bugfix	Medium	Quota pressure boundary correctly interprets stored session usage of 1.0% instead of expanding to 100%. Quota pressure boundary correctly interprets stored session usage of 1.0% instead of expanding to 100%. Source: llm_adapter@2026-05-25 Confidence: high	—

Full changelog

v9.1.1 - Claude Code Live-Path Fixes (2026-05-25)

Fixed

Read-only tool routing - prompts such as "Read notes.txt..." now use the external agent path even when initially classified as a simple query, instead of allowing a stateless model to answer without file access.
Strict-mode status banner - Claude Code session startup now reports strict zero-Claude routing when enabled, instead of misleadingly displaying subscription/MCP-tool mode.
Hook installer compatibility - global hook installation now recognizes Claude Code's nested hook settings schema and no longer crashes while reinstalling an already-registered hook.
Quota pressure boundary - a stored session usage value of 1.0% is no longer incorrectly expanded to 100% pressure.

Verified

Live Claude Code tests - validated external direct execution, fail-closed quota protection, continuation routing, and read-only file-agent execution through the installed Claude Code hook path.

View diff on GitHub

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

Share on X Share on Bluesky

Track ypollak2/llm-router

Get notified when new releases ship.

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.

All releases →

Related context

Related tools

Earlier breaking changes

v9.2.0 Changes auto‑route directive from advisory "DO NOT SKIP" to hard constraint with explicit blocked tools list.
v9.2.0 Breaks permanent downgrade of enforcement after first Edit/Write; v13 now requires per‑turn routing.