ghx

A caching proxy for the GitHub CLI (gh) that eliminates redundant API calls, prevents rate limiting, and dramatically speeds up repeated commands. Built for AI agent workflows where multiple agents (Copilot CLI, coding agents, MCP servers) hammer the same gh commands simultaneously. Highlights - 10x faster cached responses (~0.1s vs ~1s) - Singleflight coalescing — 5 agents asking the same thing = 1 API call - Allowlist-based — only caches known-safe read-only commands - Auto-invalidation — mutations flush related cache entries - Web dashboard — real-time hit rates, per-command stats, request log - Drop-in replacement — just use ghx instead of gh - No gh required — auto-downloads GitHub CLI on first use if not installed