github-deep-research
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill's primary function is to perform structured research on GitHub repositories. All network operations in the provided scripts are directed to the official GitHub API domain (api.github.com).
- [COMMAND_EXECUTION]: The skill uses a local helper script (
scripts/github_api.py) to fetch repository metadata. The script is well-structured, providing a clear CLI interface for repository analysis. It includes a fallback mechanism for therequestslibrary and securely handles authentication tokens by reading them from environment variables rather than hardcoding credentials. - [INDIRECT_PROMPT_INJECTION]: The skill represents a surface for indirect prompt injection as it ingests data from external sources such as repository READMEs, issue trackers, and web search results. However, this is inherent to the intended purpose of a research tool. The skill defines a clear output structure via
assets/report_template.mdand uses citation rules to help distinguish between external claims and agent analysis.
Audit Metadata