About Vibepod Cli
This command line tool runs AI coding agents inside isolated Docker or Podman containers from a single interface. After installing it from PyPI, Homebrew, conda forge, mamba or pixi, a developer launches an agent with one command such as vp run claude, with no configuration file required. The supported agents include Claude, Gemini, Codex, Devstral, Copilot, Auggie, OpenCode, Pi, Antigravity, Tau, Jcode, Freebuff, Qwen Code, the DeepSeek Harness and Hermes.
It is built for developers who want to try or switch between coding agents without installing each one on the host machine, and who want to see what those agents are doing. Local metrics, HTTP traffic tracking and an analytics dashboard are built in, and recent releases price each captured model call from published rates and break spend down by agent, workspace, credential profile and run. Reusable prompt recipes are available through a skills command, and a provider option routes agents to any OpenAI or Anthropic compatible endpoint or local model server. An ACP mode places containerized agents in the AI panel of editors such as Zed, PyCharm and VS Code.
Defaults are built in, and optional global or project YAML files add shared settings, custom images, environment variables, init scripts and model routing.
The tool is distributed as an open source package with its source on GitHub. The site publishes no pricing page and no paid plans.