stack-mode

Pass

Audited by Gen Agent Trust Hub on Jul 31, 2026

Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [PROMPT_INJECTION]: The skill processes untrusted content from GitHub Pull Request descriptions and comments to determine stack membership.
  • Ingestion points: The skill uses gh pr view to fetch the body and comments fields of PRs in SKILL.md.
  • Boundary markers: No explicit delimiters are used to separate external text from parsing logic; it scans raw text for markers like ## Stack.
  • Capability inventory: The skill executes gh and git CLI commands and facilitates a multi-step workflow across multiple PRs.
  • Sanitization: Pull Request numbers are extracted using a numeric regex (#[0-9]+), but the initial branch name and root reference resolution depend on broader string patterns.
  • [COMMAND_EXECUTION]: The skill includes shell script logic for interacting with local Git repositories and the GitHub CLI.
  • Evidence: Bash blocks in SKILL.md use gh pr view, gh pr list, and git branch.
  • Context: These commands are used for legitimate workflow automation to resolve stacked PR membership and are constrained by metadata extraction via regex.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 31, 2026, 01:39 AM
Security Audit — agent-trust-hub — stack-mode