resolve-greptile-comments
Pass
Audited by Gen Agent Trust Hub on Apr 8, 2026
Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes shell utilities such as grep, sed, and cut to parse pull request references from user input. It also uses the GitHub CLI (gh) to query repository information and perform GraphQL API operations for identifying and resolving specific review threads.
- [PROMPT_INJECTION]: The skill involves an indirect prompt injection surface by processing data from external GitHub pull request review threads.
- Ingestion points: Review thread IDs and comment metadata are fetched from the GitHub GraphQL API in Step 2.
- Boundary markers: No explicit delimiters or instructions are provided to the agent to ignore potentially malicious content within the fetched data.
- Capability inventory: The skill has the capability to resolve review threads via the GitHub API in Step 3.
- Sanitization: The security risk is low because the skill's logic is based on objective metadata (the bot's author login and the count of replies) rather than interpreting or executing based on the natural language content of the comments.
Audit Metadata