agent-md-setup
Pass
Audited by Gen Agent Trust Hub on Sep 4, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill's primary function is to manage local documentation files (AGENTS.md and CLAUDE.md) within a repository using provided templates.
- [COMMAND_EXECUTION]: The skill uses
git statusto inspect the repository state, which is a benign and standard development operation. - [INDIRECT_PROMPT_INJECTION]: The skill ingests existing AGENTS.md and CLAUDE.md files which could contain malicious instructions. However, the skill explicitly mandates that the agent must 'Summarize what exists and what would change before writing' and 'make no change until the user approves it,' providing a robust human-in-the-loop check against automated exploitation.
- [DATA_EXPOSURE]: The skill only interacts with non-sensitive repository documentation files. It does not access credentials, private keys, or environment variables.
Audit Metadata