review-contract

Pass

Audited by Gen Agent Trust Hub on Sep 11, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from external sources, specifically local PDF/docx files and documents from EDO systems (Diadoc/Sbis). This represents a vulnerability surface where a malicious document could contain embedded instructions intended to override the agent's behavior.
  • Ingestion points: The skill accepts local file paths and document IDs from EDO platforms as defined in SKILL.md (Step 1).
  • Boundary markers: The instructions do not define specific delimiters or "ignore instructions" warnings to isolate the contract text from the agent's system instructions.
  • Capability inventory: The skill utilizes Read for file access, WebFetch for external EDO connectivity, and Bash for exporting files, which could be leveraged if an injection succeeds.
  • Sanitization: There is no evidence of content sanitization or validation of the text extracted from the contracts.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 11, 2026, 04:25 PM
Security Audit — agent-trust-hub — review-contract