contractor-agreement-review
Pass
Audited by Gen Agent Trust Hub on Sep 30, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted contractor agreements uploaded by the user, creating a surface for potential indirect prompt injection attacks.
- Ingestion points: The
SKILL.mdfile defines workflows for analyzing user-uploaded documents inCHECKandADVISEmodes. - Boundary markers: The instructions do not define specific delimiters or instructions to ignore embedded commands within the analyzed document.
- Capability inventory: The skill is limited to text analysis and reporting; it lacks any capabilities for command execution, file writing, or network operations beyond outputting a static URL.
- Sanitization: No content sanitization or validation is specified for the input document.
- [SAFE]: The skill performs legitimate text analysis for contract review using provided reference files. No indicators of obfuscation, exfiltration, or unauthorized code execution were found.
Audit Metadata