Ross ROSS = Recommend OSS · open-source software intelligence for agents

Claude Code Router

Some data may be outdated — see provenance labels.

One local control plane for every AI agent: route across models, fuse new capabilities, orchestrate tools, and stay fully in control. observed · 2026-09-22

github.com/musistudio/claude-code-router · homepage · TypeScript · MIT (permissive) observed · 2026-09-22

Health v2 · maintenance only

85/100

  • Activity 95
  • Release rhythm 95
  • Longevity 43
How is this computed?

round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10) — computed 2026-10-21. Adoption (stars, forks) is never an input.

  • gap_med: 1.5
  • age_days: 603
  • days_rel: 34
  • days_push: 30
  • n_releases_24m: 25

Full methodology

Adoption not part of the score

37369 stars · 3141 forks observed · 2026-09-22

What it is AI-extracted, prompt v1, taxonomy v1, 2026-09-28, confidence 0.97

Claude Code Router is a local control plane that routes requests from AI agents to multiple LLM providers and can fuse capabilities like vision or search. It runs as a desktop app, CLI tool, or Docker service and offers logging, observability, and fail‑over features.

Use cases

  • Route Claude Code or other coding agents to the best LLM provider based on custom rules
  • Combine text models with vision, search, or custom tools to create a new AI capability
  • Fail‑over between multiple API keys or providers when a request fails
  • Expose a local AI gateway that can be accessed from Slack, Discord, or other IM platforms
  • Monitor token usage, latency, and cost for every model request in a single dashboard

When to choose

  • You need a self‑hosted gateway to control and audit LLM usage across multiple providers
  • You want to experiment with model fusion or custom routing logic without writing code
  • You run AI agents locally and require a simple CLI or desktop UI for configuration
  • You need detailed logs and observability for compliance or cost tracking

When to avoid

  • You only use a single provider and do not need routing or fusion features
  • You require a cloud‑native, fully managed service with zero maintenance overhead
  • Your environment cannot run Node.js or Docker (e.g., restricted embedded systems)
  • You need real‑time low‑latency inference that cannot tolerate an additional gateway layer

Facets

application · maturity active

api-gateway routing llm-inference cli large-language-models windows cli local-llm-gateway model-fusion toolhub agentclaw multi-provider tool-integration ai-agents retrieval-augmented-generation devops linux macos docker nodejs desktop

3 sources

Member repositories

RepositoryRoleHealth v2
musistudio/claude-code-routermain85
musistudio/llmsbackend43

For agents

markdown · JSON · MCP: product_card(name="musistudio/claude-code-router")

Data as of 2026-10-19T08:05:51.212177+00:00 · Report a problem