Run Agent

This is your Swiss Army knife for running AI agents across different LLM providers without rewriting code. It handles Claude, OpenAI, CrewAI, Lyzr, and GitHub Models through a unified interface, so you can swap adapters with a flag instead of refactoring. The git repo support is genuinely handy: point it at a repository and it'll clone, cache, and even auto-detect which adapter to use based on config files or model names. One-shot mode lets you fire off prompts without interactive sessions, and the caching system means you're not re-cloning repos every time. If you're experimenting with different LLM providers or maintaining agents that need to run on various platforms, this saves you from maintaining separate CLI wrappers for each one.

GitHub stars: 2.8K
last updated: May 28, 2026

View on GitHub

Install

$ npx -y skills add open-gitagent/opengap --skill run-agent --agent claude-code

Installs into .claude/skills of the current project.

Files

Run Agents

When to Use

When a user wants to run an agent locally, from a git repo, or with a specific adapter/framework.

Troubleshooting

If you see authentication errors:

Claude: Ensure Claude Code is authenticated (claude auth status)
OpenAI: Verify OPENAI_API_KEY is set and valid
GitHub: Check GITHUB_TOKEN has correct permissions
Lyzr: Confirm LYZR_API_KEY is active

Basic Usage

# Run local agent with Claude (default)
opengap run -d ./my-agent

# Run from git repo
opengap run -r https://github.com/user/agent

# Run with a prompt (one-shot mode)
opengap run -d ./my-agent -p "Review my code"