plan-eng-review

Pass

Audited by Gen Agent Trust Hub on Mar 13, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: No malicious patterns or behaviors detected. The skill performs architectural and code reviews as described using natural language instructions.\n- [PROMPT_INJECTION]: The skill instructions define a surface for processing untrusted data by reading repository-local instruction files, which is a potential vector for indirect prompt injection.\n
  • Ingestion points: The skill reads AGENTS.md, TODO.md, and TODOS.md for local repository instructions.\n
  • Boundary markers: Absent; the skill does not specify delimiters or instructions to isolate or ignore embedded commands within these external files.\n
  • Capability inventory: The skill is designed for review and documentation purposes; it does not explicitly invoke shell commands, network operations, or file system writes that could be abused through injection.\n
  • Sanitization: No sanitization or validation logic for the content of ingested files is defined.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 13, 2026, 08:29 PM