domain-language

Pass

Audited by Gen Agent Trust Hub on Sep 12, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill is designed to read from a 'canonical domain glossary' to guide its vocabulary usage. This introduces an attack surface where malicious instructions could be embedded within the glossary content to influence the agent's behavior.
  • Ingestion points: The skill reads external glossary files during the 'Read' step in SKILL.md.
  • Boundary markers: There are no specific boundary markers or instructions to ignore potential commands within the glossary text.
  • Capability inventory: The skill has file-read and file-write capabilities to update the glossary. It does not exhibit network or shell execution capabilities.
  • Sanitization: No sanitization or validation of the glossary content is described, meaning the agent treats the content as authoritative vocabulary guidelines.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 12, 2026, 11:53 PM
Security Audit — agent-trust-hub — domain-language