skills/greptileai/skills/check-pr/Gen Agent Trust Hub

check-pr

Pass

Audited by Gen Agent Trust Hub on Sep 14, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from pull request descriptions and comments, which could potentially contain malicious instructions.
  • Ingestion points: Fetches PR/MR bodies and discussion threads using gh pr view, glab api, and p4 describe (SKILL.md).
  • Boundary markers: No specific delimiters or "ignore" instructions are used to separate user-provided content from agent instructions.
  • Capability inventory: The skill is capable of performing write operations such as git push, p4 shelve, and resolving review threads via API calls (SKILL.md, Step 7 and 8).
  • Sanitization: The instructions do not define any sanitization or filtering logic for the fetched content before it is processed by the agent.
  • [COMMAND_EXECUTION]: The skill constructs shell commands for GitHub, GitLab, and Perforce using variables like PR numbers and branch names. While these are necessary for the skill's operation, they represent a potential attack surface if the execution environment does not properly validate these identifiers.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 14, 2026, 07:02 PM
Security Audit — agent-trust-hub — check-pr