L

llm-gateway

Put a queue, deduplication, a circuit breaker and a dead-letter queue in front of any LLM API (Anthropic, OpenAI, llama.cpp, vLLM), so repeated prompts cost one call and outages fail fast. Rust library, HTTP server and CLI.

An LLM gateway that can prove what happened. One OpenAI-compatible endpoint in front of AWS Bedrock and on-device models, with per-team cost attribution, redaction no application can bypass, and a tamper-evident record of every completion.

Working notes for pointing the Anthropic Go SDK at an Anthropic-compatible gateway through ANTHROPIC_BASE_URL (Antigravity proxy), plus the glab CLI issue-and-branch workflow shared across these agent projects. Specification and prompts only - no implementation committed yet.