legal-advisor

Pass

Audited by Gen Agent Trust Hub on Aug 27, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill is designed to process external user data and web content to generate legal documents, which represents an ingestion point for untrusted data.
  • Ingestion points: Processes user goals, constraints, and web content (via BrowserOS) in SKILL.md.
  • Boundary markers: The instructions include a specific 'Security & Multi-Agent Hygiene' section that mandates treating external data as hostile.
  • Capability inventory: The agent uses MemoryStore Interface for file writes and ExecutionProxy Interface for terminal commands.
  • Sanitization: Explicitly requires redacting secrets and PII before sharing context with subagents.
  • [COMMAND_EXECUTION]: The skill defines a protocol for executing terminal actions using a specific proxy interface (e.g., rtk npm test). This is a description of the agent's framework capabilities rather than a specific malicious command.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 27, 2026, 07:35 AM
Security Audit — agent-trust-hub — legal-advisor