git-issue-hierarchy

Pass

Audited by Gen Agent Trust Hub on Apr 21, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill uses the official GitHub CLI (gh) to interact with repository data, facilitating legitimate issue management workflows.
  • [SAFE]: Employs dynamic context injection (!git remote get-url origin) to resolve the repository URL at load time, which is a benign and standard use case for repository-aware tooling.
  • [SAFE]: Implements restrictive tool access via allowed-tools, limiting shell execution to specific gh and git subcommands, which follows the principle of least privilege.
  • [PROMPT_INJECTION]: The skill identifies an ingestion surface by parsing issue bodies for dependency markers like 'Blocked by #N'. While this involves processing untrusted external content, the logic is limited to building a dependency tree and does not interpret the content as executable instructions.
  • Ingestion points: Issue bodies and titles retrieved via 'gh issue view' command.
  • Boundary markers: None present.
  • Capability inventory: Issue creation and modification via 'gh issue' and 'gh api' calls.
  • Sanitization: Not explicitly implemented as the parsed content is treated as metadata.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 21, 2026, 01:18 AM