AI Briefingrouter: A local proxy that automatically selects the optimal model for each client request
workweave/router
About the project
Connect various AI coding tools such as Claude Code, Codex, and Cursor through a single localhost:8080 endpoint. Manage Anthropic, OpenAI, Gemini, as well as open-source models like DeepSeek and Llama through one unified endpoint.
The core feature is routing at the 'action' level. A built-in embedder analyzes request content to automatically assign the most suitable model. Rather than relying on simple prompt-based guessing, it optimizes performance and cost based on the actual task context.
Ideal for developers who prioritize data privacy. API keys are encrypted and stored locally, and prompts are sent directly to providers without passing through Weave servers. It supports OTLP tracing, enabling integration with existing monitoring tools such as Honeycomb and Datadog.
workweave/router
Model router for agentic systems. Routes every prompt to the right model in <50ms. Cut costs 40-70% with just an endpoint change.
Go
This introduction was generated automatically by AI. Check the original for the author's claims and context. Copyright belongs to the original author.
Our guide explains how the AI works. Report errors, attribution issues, or removal requests via Contact.