This skill provides a rigorous framework for creating run-<unit> skills. Instead of simply paraphrasing READMEs, it forces the agent to actually execute the software, build a programmatic interaction harness (driver), and verify the result through screenshots or logs before documenting the process.
Activate this skill when you need to set up a project for the first time, create instructions for future agents to launch and test the app, or verify that a project's build and run steps are fully operational in a clean environment.
SKILL.md including Prerequisites, Build, Agent Path, and Gotchas.Specifically tailored for Claude Code and similar agents that support local tool execution, PTYs (tmux), and browser automation (chromium-cli).
This skill has not been reviewed by our automated audit pipeline yet.