CLI AI Proxy

Local OpenAI-compatible proxy that bridges Gemini CLI and Claude Code to a unified REST API. Requests go through the installed CLI tools — the proxy makes no direct AI-vendor API calls and holds no API keys. Authentication (OAuth, API keys, session tokens) is managed by the gemini / claude CLIs themselves.

What This Installs

This skill installs the cli-ai-proxy package from the public npm registry. Specifically:

• Source: public npm registry package cli-ai-proxy (no GitHub clone, no local build step)
• Postinstall scripts: none — the package's package.json declares no preinstall/postinstall hooks
• Runtime dependencies: one — yaml (config parsing)
• Filesystem writes: only under the global npm prefix (for the binary) and, when the proxy runs, under its working directory for config.yaml, .proxy.pid, and proxy.log
• Config mutations: only if you explicitly run configure-provider.sh / cli-ai-proxy configure-openclaw, which edits ~/.openclaw/openclaw.json and writes a .bak backup next to it first
• Network at runtime: localhost HTTP server only; outbound calls are performed by the user's installed gemini / claude CLIs, not by the proxy itself

When to Use

✅ User asks to start/stop/check the AI proxy ✅ User wants to route requests through Gemini CLI or Claude Code ✅ User asks about available models or proxy health ✅ User wants to configure OpenClaw to use the proxy ✅ Troubleshooting proxy connectivity or CLI issues

❌ Direct API calls to OpenAI/Anthropic/Google (this proxy only uses CLI tools) ❌ Managing API keys (CLIs handle their own authentication)

Quick Reference

Proxy Lifecycle

Starting

{baseDir}/scripts/start.sh

Starts the proxy on 127.0.0.1:9090 (default). The proxy listens for OpenAI-compatible requests and routes them to the appropriate CLI tool.

Before starting, verify at least one CLI is available:

• gemini --version (Gemini CLI)
• claude --version (Claude Code)

Checking Status

{baseDir}/scripts/status.sh

Shows: running/stopped, PID, health endpoint data, available CLI providers, concurrency stats.

Stopping

{baseDir}/scripts/stop.sh

Gracefully shuts down the proxy: stops accepting connections, kills active CLI subprocesses, cleans up.

Available Models

When OpenClaw is configured, use as cli-ai-proxy/gemini, cli-ai-proxy/claude, etc.

OpenClaw Integration

To configure OpenClaw to route through the proxy:

{baseDir}/scripts/configure-provider.sh

This automatically:

1. Adds cli-ai-proxy as a provider in ~/.openclaw/openclaw.json
2. Registers all proxy models in the agent defaults
3. Creates a backup of the original config

After configuring, set the default model in openclaw.json:

{ "agents": { "defaults": { "model": { "primary": "cli-ai-proxy/gemini" } } } }

API Endpoints

The proxy exposes:

• POST /v1/chat/completions — Chat completions (streaming + non-streaming)
• GET /v1/models — List available models
• GET /health — Health check with provider status and concurrency info

Default base URL: http://127.0.0.1:9090/v1

For full API details see references/api.md.

Image Support

The proxy supports images in messages. When a request contains image_url content parts:

1. Images are saved to temporary files
2. The prompt instructs the CLI to read the image via its built-in file tools
3. Temp files are automatically cleaned up after each request

Supports both base64 data URLs and remote image URLs.

Configuration

Config file: config.yaml in the proxy installation directory.

Key settings:

• server.port — Listen port (default: 9090)
• concurrency.max — Max concurrent CLI processes (default: 5)
• timeout — CLI process timeout in ms (default: 300000)
• defaultModel — Default model when none specified

For full configuration options see references/configuration.md.

Troubleshooting

Proxy won't start

1. Check if port 9090 is already in use: lsof -i :9090
2. Verify Node.js is available: node --version
3. Check logs: read the proxy.log file in the installation directory

CLI not available

1. Verify CLI is installed and in PATH: which gemini or which claude
2. Check CLI auth: gemini --version or claude --version
3. The proxy health endpoint shows which CLIs are available

429 Too Many Requests

The concurrency limit has been reached. Either:

• Wait for current requests to complete
• Increase concurrency.max in config.yaml

Timeout errors (504)

The CLI process took too long. Either:

• Increase timeout in config.yaml
• Check if the CLI is hanging (auth issues, network)

For more troubleshooting see references/troubleshooting.md.