legal-contract-review
Pass
Audited by Gen Agent Trust Hub on Sep 14, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and analyze external, untrusted legal documents (such as contracts, terms of service, and NDAs) provided by the user, which could potentially contain malicious instructions intended to influence the agent's behavior.
- Ingestion points: The skill accepts user-provided text and files (PDF, DOCX, images) for analysis as described in SKILL.md and references/contract/guide.md.
- Boundary markers: The skill includes instructions to mask personally identifiable information (PII) using the 'moai-coworker:ai-slop-reviewer' skill and provides explicit legal disclaimers to limit the scope of the AI's output.
- Capability inventory: The skill's primary capabilities are text analysis and document generation. No high-risk capabilities such as arbitrary command execution, network exfiltration, or unauthorized file system modification were detected in the provided scripts.
- Sanitization: While PII masking is implemented, there are no specific filters or sanitization steps mentioned to detect or neutralize embedded natural language instructions within the documents being analyzed.
Audit Metadata