ubiquitous-language

Pass

Audited by Gen Agent Trust Hub on Sep 7, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill is designed to scan user-provided conversation content to extract domain terms and definitions. This creates an attack surface where malicious instructions embedded in the conversation (e.g., in a chat transcript provided by the user) could attempt to influence the agent's behavior.
  • Ingestion points: Reads the 'current conversation' (SKILL.md).
  • Boundary markers: None defined. The skill lacks explicit instructions to ignore embedded commands within the conversation text.
  • Capability inventory: File write access to UBIQUITOUS_LANGUAGE.md (SKILL.md).
  • Sanitization: None detected. The skill directly interpolates extracted concepts into the markdown output.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 7, 2026, 02:48 PM
Security Audit — agent-trust-hub — ubiquitous-language