missing-protections

Pass

Audited by Gen Agent Trust Hub on Sep 9, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external data (contract files) using the Read tool to perform a legal gap analysis. There is a potential risk of indirect prompt injection if a contract contains hidden instructions intended to manipulate the agent's behavior or override its auditing logic.
  • Ingestion points: Full contract text ingested via the Read tool from user-specified file paths.
  • Boundary markers: The instructions do not specify the use of delimiters or explicit "ignore embedded instructions" warnings for the ingested text.
  • Capability inventory: The skill uses filesystem tools (Read, Glob, Grep) and generates a markdown report file based on the analysis.
  • Sanitization: No explicit sanitization or validation of the contract content is mentioned before it is interpolated into the agent's context.
  • [EXTERNAL_DOWNLOADS]: The skill provides links to external resources for legal templates and regulatory guidance, including CommonPaper, the Information Commissioner's Office (ICO), and official GDPR documentation. These are well-known and reputable sources for the skill's specific domain of legal auditing.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 9, 2026, 03:48 AM
Security Audit — agent-trust-hub — missing-protections