janitor
Pass
Audited by Gen Agent Trust Hub on Sep 13, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill is designed to execute shell commands using the git and gh (GitHub CLI) tools to perform audits, cleanup, and commit management. While these are necessary for the skill's function, they provide a broad capability to read and modify the filesystem and remote repository state. The skill includes safety rails requiring explicit user authorization for all destructive operations, such as force-pushing or branch deletion.
- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from repository history, staged diffs, and GitHub metadata which are processed by the agent to generate messages and reports.
- Ingestion points: Commit messages (git log), staged changes (git diff), and GitHub API data (gh pr list, gh issue list, gh repo view).
- Boundary markers: No explicit delimiters or instructions are provided in the skill files to isolate untrusted content from the agent's instructions.
- Capability inventory: The skill possesses access to the git and gh CLI tools, allowing for filesystem reading/writing and network access to GitHub.
- Sanitization: No explicit sanitization or filtering of external data from PRs or commit messages is defined.
Audit Metadata