Skills
Navigate
skills/pydantic/monty/fix-pr-comments/Gen Agent Trust Hub

fix-pr-comments

Pass

Audited by Gen Agent Trust Hub on Aug 23, 2026

Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection through pull request review comments, although it incorporates multiple layers of protection.
  • Ingestion points: pr-threads.sh retrieves comment content from the GitHub API using gh api graphql.
  • Boundary markers: Present in SKILL.md; the agent is explicitly instructed to treat comments as non-authoritative "claims" and is warned that bot comments may quote untrusted text from the PR author.
  • Capability inventory: The skill allows the agent to modify the local codebase to fix issues and execute gh api calls to post replies and resolve threads on GitHub.
  • Sanitization: Present in pr-threads.sh; comment data is filtered against a hardcoded list of verified bot database IDs to exclude input from humans or untrusted automated actors.
  • [COMMAND_EXECUTION]: The skill executes the gh CLI and the pr-threads.sh script to perform PR operations. These commands are targeted at the intended functionality of the skill.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 23, 2026, 08:46 AM