ubiquitous-language

Pass

Audited by Gen Agent Trust Hub on Jul 31, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill ingests and processes untrusted conversation history to generate a persistent file, creating a surface for indirect prompt injection.
  • Ingestion points: The skill scans nouns, verbs, and concepts from the conversation history.
  • Boundary markers: The instructions lack explicit delimiters or warnings to ignore commands embedded in the conversation data.
  • Capability inventory: The skill is authorized to read and write a local file named UBIQUITOUS_LANGUAGE.md and output summaries to the conversation.
  • Sanitization: No sanitization or escaping of the user-provided conversation text is specified before file writing.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 31, 2026, 08:38 AM
Security Audit — agent-trust-hub — ubiquitous-language