write-agents-md
Pass
Audited by Gen Agent Trust Hub on Oct 18, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill is designed with a strong focus on security and integrity. It uses a 'copy-first' approach with a dedicated helper script to ensure consistency and prevent accidental errors or malicious modifications.
- [COMMAND_EXECUTION]: The skill requires running a local Python script
scripts/copy_template.py. This script is benign and contains the following security features: 1. Path Validation: It usesPath.is_relative_toto ensure that files are only written within the intended project root, preventing directory traversal attacks. 2. Overwrite Protection: It uses the exclusive creation mode (xb) to prevent accidentally overwriting existing project files. 3. Integrity Verification: It calculates and reports a SHA-256 hash of the copied template and performs a byte-level comparison to ensure the copy is exact. - [DATA_EXPOSURE]: The instructions throughout the skill (e.g., in
SKILL.mdandreferences/endpoint-mocks.md) explicitly remind the user to never commit private credentials, tokens, or sensitive data to the repository, advocating for standard security hygiene.
Audit Metadata