ypollak2/llm-router

v1.8.3 Bugfix

This release fixes issues for SREs watching stability and regressions.

Published 1mo 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

Affected surfaces

breaking_upgrade

Summary

AI summary

Fixed critical routing failures in CLI and agent mode caused by hook format drift and missing MCP server registration.

Full changelog

Critical bug fixes

Two bugs that caused complete routing failure in CLI and agent mode (zero cost savings, all tasks hit Opus).

Bug 1: Hook format drift — routing silently ignored

The auto-route.py hook was emitting ⚡ ROUTE→tool(...) but the rules file told Claude to look for ⚡ MANDATORY ROUTE:. Claude saw no matching directive and ignored routing entirely.

Fix: hook now emits the canonical ⚡ MANDATORY ROUTE: {task_type}/{complexity} → call {tool}({args}) format.

Bug 2: MCP server not visible to `claude -p`

llm-router install was writing mcpServers only to ~/.claude/settings.json (Claude Desktop), not ~/.claude.json (Claude Code CLI). After a fresh install, claude mcp list showed no llm-router entry.

Fix: install() now also registers via claude mcp add --scope user (with a direct JSON merge fallback for Docker/headless environments).

Also fixed

CI publish workflow: added timeout-minutes: 15, skip_existing: true, explicit --timeout=30

Upgrade

pip install --upgrade claude-code-llm-router
llm-router install --force  # re-installs hooks and re-registers MCP

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.