security-audit

Pass

Audited by Gen Agent Trust Hub on May 10, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: Uses dynamic context injection to execute shell commands (git rev-parse, git diff, git log, gh pr view) to gather repository state at load time.\n- [PROMPT_INJECTION]: The skill ingests untrusted pull request metadata which introduces an indirect prompt injection surface.\n
  • Ingestion points: Pull request title, body, and comments are fetched via gh pr view and injected into the orientation context in SKILL.md.\n
  • Boundary markers: Absent; metadata is interpolated directly into the instructions as bullet points without delimiters.\n
  • Capability inventory: The skill dispatches a separate security-auditor agent that performs vulnerability audits on the branch code.\n
  • Sanitization: No validation or sanitization is performed on the external PR content before injection.
Audit Metadata
Risk Level
SAFE
Analyzed
May 10, 2026, 07:59 AM
Security Audit — agent-trust-hub — security-audit