terms-generator

Pass

Audited by Gen Agent Trust Hub on Jul 31, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection because it ingests and processes untrusted data from external websites and local codebases which could contain hidden malicious instructions.\n
  • Ingestion points: The skill uses the WebFetch tool to scan user-provided URLs and the Read, Glob, and Grep tools to scan local project code (SKILL.md, instructions 1 and 2).\n
  • Boundary markers: The instructions do not specify the use of delimiters or 'ignore' instructions to separate the untrusted data from the agent's primary instructions.\n
  • Capability inventory: The skill possesses the Write tool, allowing it to create or modify files on the local filesystem based on the analysis of untrusted data.\n
  • Sanitization: There are no instructions to sanitize, filter, or escape content retrieved from the web or local files before it is processed by the agent.\n- [EXTERNAL_DOWNLOADS]: The skill references several well-known and official regulatory resources for compliance guidance, such as the Federal Trade Commission (FTC), the California Attorney General, and the UK Information Commissioner's Office (ICO).
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 31, 2026, 02:22 PM
Security Audit — agent-trust-hub — terms-generator