Minimal LLM API reverse proxy for Claude Code, Codex CLI & Gemini CLI . YAML routing, hot reload, and automatic provider failover. 极简 LLM API 反向代理:面向 Claude Code / Codex CLI / Gemini CLI,支持 YAML 配置、多上游优先级与自动故障转移。
git clone https://github.com/lansespirit/Clipal.gitClipal is a local LLM API proxy designed for developers using AI coding tools like Claude Code, Continue, Aider, and Gemini CLI. It consolidates multiple model providers into a single endpoint at http://127.0.0.1:3333, automatically rotating between API keys and failing over to secondary providers when rate limits or quota issues occur. The Web UI enables one-click CLI takeover to configure supported tools without manual file editing, plus OAuth support for Anthropic, OpenAI, and Google accounts. Clipal runs as a cross-platform background service on macOS, Linux, and Windows, with hot-reload configuration changes and automatic access token refresh.
Install Clipal via npm (npm install -g clipal) or download a standalone binary from GitHub Releases. Run clipal to start the proxy and open http://127.0.0.1:3333 in your browser. Use the Web UI to add providers, configure API keys or OAuth accounts, then click CLI Takeover to automatically configure your coding tool. Optionally install as a background service with clipal service install.
Consolidate multiple LLM providers and API keys into one local endpoint
Automatically rotate between API keys to avoid rate limits during long coding sessions
Fail over to backup providers or models when primary quota is exhausted
Configure Claude Code, Codex CLI, Continue, and Aider with a single Web UI click
No install command available. Check the GitHub repository for manual installation instructions.
git clone https://github.com/lansespirit/ClipalCopy the install command above and run it in your terminal.
Launch Claude Code, Cursor, or your preferred AI coding agent.
Use the prompt template or examples below to test the skill.
Adapt the skill to your specific use case and workflow.
Set up a Clipal reverse proxy for [COMPANY] to route requests between [PRIMARY_PROVIDER] and [FALLBACK_PROVIDER]. Use the following YAML configuration: [YAML_CONFIG]. Ensure automatic failover is enabled and hot reload is configured for [INDUSTRY] compliance.
```yaml
# Clipal Configuration for AI Model Routing
providers:
- name: Claude Code
api_key: ${CLAUDE_API_KEY}
priority: 1
- name: Gemini CLI
api_key: ${GEMINI_API_KEY}
priority: 2
routes:
- path: /api/v1/chat
provider: Claude Code
- path: /api/v1/code
provider: Gemini CLI
failover:
enabled: true
max_retries: 3
retry_delay: 1000
reload:
enabled: true
interval: 30000
```
### Configuration Explanation
- **Providers**: Defines the primary and fallback providers with their respective API keys and priorities.
- **Routes**: Specifies the paths and the corresponding providers for each route.
- **Failover**: Enables automatic failover with a maximum of 3 retries and a 1-second delay between retries.
- **Reload**: Enables hot reload with a 30-second interval for configuration updates.Google's multimodal AI model and assistant
AI assistant built for thoughtful, nuanced conversation
Automate your browser workflows effortlessly
Get more done every day with Microsoft Teams – powered by AI
Agentic AI Workflow platform
Connected workspace for docs, wikis, and projects
Take a free 3-minute scan and get personalized AI skill recommendations.
Take free scan