#224224th of 489 in Skills, plugins and prompts
copilot-api
Local gateway exposing Copilot, Codex, xAI and other providers via OpenAI/Anthropic APIs
- Stars
- 1.1k
- License
- MIT
- Last commit
- Nov 2026
- Last release
- Nov 2026
- Language
- TypeScript
654 on 25 Apr 20261,129 on 14 Nov 2026
From GitHub's daily star counts.
Checked on GitHub.
Overview
Runs a local server on port 4141 that serves Chat Completions, OpenAI Responses and Anthropic Messages endpoints, and routes requests to GitHub Copilot, built-in Codex and xAI providers, or third-party providers such as Kimi, DeepSeek and OpenRouter. It translates between protocols so Claude Code, OpenCode and Codex can use the same upstream. An Electron desktop app and token usage monitoring are included.
Who it is for: Developers pointing Claude Code, OpenCode or Codex at Copilot or other providers
Strengths
- Three client protocols (Chat Completions, Responses, Messages) from one endpoint, with adapters between them
- Provider-only mode works without a GitHub token or Copilot subscription
- Per-client setup guides for Claude Code, OpenCode, Codex and DeepSeek Harness
- Electron desktop app for sign-in, provider setup, usage and logs
Weaknesses
- Protocol support is model-specific; Chat Completions needs a native upstream endpoint
- Using Copilot as upstream carries a security notice the README tells you to read first
- Runtime CVE-2026-111497 requires patched Node.js or Bun versions to be installed
- Desktop installers cover only Windows x64, macOS Apple Silicon and Linux x64
What it needs
- no GPU
- Docker + Compose
- Needs Node.js 22/24/26 or Bun >= 1.4.4, GitHub Copilot subscription (only for the Copilot provider)
- Models: GitHub Copilot, OpenAI Codex, xAI, Kimi, DeepSeek
- port 4141
What changed
From the release notes. No flag means the note does not say so, not that the upgrade is safe.
Also in Skills, plugins and prompts
See all 489| Rank | Project | Score |
|---|---|---|
| 1 | claude-memPersistent cross-session memory plugin for Claude Code and other coding agents | 97 out of 100 |
| 2 | impeccableDesign skill with 24 slash commands for AI coding agents' frontend output | 97 out of 100 |
| 3 | gstackMarkdown slash-command skills that give Claude Code planning, review, QA and ship roles | 96 out of 100 |
| #4 | rufloAgent harness adding swarms, memory and hooks to Claude Code and Codex | 96 out of 100 |
| #5 | eccAgents, skills, hooks and memory bundle for Claude Code and other harnesses | 95 out of 100 |