00-andruia-consultant

Pass

Audited by Gen Agent Trust Hub on Sep 6, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill instructions are limited to project documentation, diagnostic interviews, and planning. It does not contain any code execution, network requests, or sensitive data access.
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data by scanning existing project files (e.g., src, package.json). Evidence Chain: 1. Ingestion points: Local workspace files detected in Scenario B of SKILL.md. 2. Boundary markers: Absent. 3. Capability inventory: File reading and creation of documentation files (.md). 4. Sanitization: None specified. The potential impact is low as the agent's capabilities are restricted to generation of planning artifacts.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 6, 2026, 12:45 AM
Security Audit — agent-trust-hub — 00-andruia-consultant