github-ai-trends-hardened
Pass
Audited by Gen Agent Trust Hub on Apr 21, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill incorporates comprehensive security instructions that define strict boundaries for the agent, including prohibitions on disk persistence, credential logging, and autonomous error recovery.
- [EXTERNAL_DOWNLOADS]: Accesses the official GitHub API (api.github.com) to fetch repository metadata. This is a well-known service and the data fetched is used exclusively for generating the requested report.
- [COMMAND_EXECUTION]: Strictly limits command execution to a single local Python script (scripts/fetch_trends.py). It specifically forbids the use of other shell tools or utilities like curl or git.
- [SAFE]: Properly manages the GITHUB_TOKEN by encouraging environment variable usage and mandating that the token never be displayed or leaked in chat outputs.
Audit Metadata