gh-review-requests

Pass

Audited by Gen Agent Trust Hub on Sep 15, 2026

Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The Python script scripts/fetch_review_requests.py uses subprocess.run to call the GitHub CLI (gh). The implementation correctly passes commands as a list rather than a single string, which prevents shell command injection vulnerabilities even when handling user-provided team slugs.
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from GitHub (PR titles and author names) and presents them to the user in a markdown table.
  • Ingestion points: GitHub API responses retrieved via gh api in scripts/fetch_review_requests.py.
  • Boundary markers: The skill does not currently use explicit delimiters or "ignore instructions" warnings when rendering the PR titles.
  • Capability inventory: The skill utilizes Bash and the GitHub CLI to read data; it does not perform write operations or execute code based on the PR content.
  • Sanitization: PR metadata is displayed directly in a markdown table without sanitization. While this is a theoretical surface for injection, the impact is minimized by the agent's restricted scope of displaying the data.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 15, 2026, 08:28 AM
Security Audit — agent-trust-hub — gh-review-requests