The Agent Skills Directory

[SAFE]: The skill is designed as a read-only routing utility. It does not perform network operations, exfiltrate data, or attempt to modify the underlying system.
[INDIRECT_PROMPT_INJECTION]: The skill analyzes external repository data (such as CONTEXT.md, AGENTS.md, and package.json) which could theoretically contain malicious instructions. However, it explicitly mitigates this risk by instructing the agent to treat repository text as project data rather than authoritative instructions.
[COMMAND_EXECUTION]: The skill employs benign, read-only commands such as git status and git diff to determine the current state of the repository, which is consistent with its intended function.

ask-workflow