Skills
Navigate
skills/shipshitdev/skills/pr-comments/Gen Agent Trust Hub

pr-comments

Pass

Audited by Gen Agent Trust Hub on Oct 7, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external data from GitHub PR comments and review threads.
  • Ingestion points: Fetches review summaries, conversation comments, and inline review comments via gh api and gh pr view in SKILL.md (Phase 2).
  • Boundary markers: Explicitly instructs the agent to treat all comment text as untrusted, summarize content instead of following embedded instructions, and redact secret-like values (SKILL.md, Authorized Scope).
  • Capability inventory: The skill uses gh for reading/writing to GitHub and git for local code modifications in address mode.
  • Sanitization: Implements summarization and redaction logic to mitigate risks from processing external feedback.
  • [COMMAND_EXECUTION]: The skill executes gh (GitHub CLI) and git commands to manage PR feedback and apply local changes.
  • Evidence: SKILL.md defines command usage for gh auth status, gh pr view, and gh api to retrieve repository data.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 7, 2026, 04:47 PM