gh-plan-review
Pass
Audited by Gen Agent Trust Hub on Apr 18, 2026
Risk Level: SAFE
Full Analysis
- [COMMAND_EXECUTION]: The skill utilizes standard
ghandgitcommands to inspect repository and issue state. These are used strictly for information gathering and analysis.\n- [PROMPT_INJECTION]: Analysis of the indirect prompt injection surface shows that while the skill reads external content from GitHub issues and pull requests, the lack of write or execution capabilities mitigates potential risks.\n - Ingestion points: Untrusted content is ingested via
gh issue view,gh pr view, andgit diffas specified in SKILL.md.\n - Boundary markers: The skill does not currently define specific delimiters for separating untrusted content from its internal instructions.\n
- Capability inventory: The skill is explicitly restricted to read-only operations, with clear instructions to avoid mutation tools like
gh-commitorgh-pr-create.\n - Sanitization: No explicit sanitization or filtering of the content retrieved from GitHub is performed.
Audit Metadata