contract-review-risk-scoring

Pass

Audited by Gen Agent Trust Hub on Jun 16, 2026

Risk Level: SAFE
Full Analysis
  • [PROMPT_INJECTION]: The skill defines a process for analyzing untrusted contract text provided by users, which is a potential surface for indirect prompt injection.\n
  • Ingestion points: The workflow depends on users pasting full contracts (NDAs, SaaS agreements, etc.) into the context as defined in SKILL.md.\n
  • Boundary markers: The instructions do not define specific delimiters or guards to prevent the agent from potentially following instructions found within the contract text.\n
  • Capability inventory: No tools, subprocesses, or network access capabilities are defined in the skill.\n
  • Sanitization: There is no evidence of input validation or content filtering for the processed documents.\n- [EXTERNAL_DOWNLOADS]: The skill metadata and documentation reference an external GitHub repository (evolsb/claude-legal-skill) for installation purposes, but the skill itself does not perform any automated downloads or remote code execution at runtime.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 16, 2026, 07:54 PM
Security Audit — agent-trust-hub — contract-review-risk-scoring