ubiquitous-language

Pass

Audited by Gen Agent Trust Hub on Aug 12, 2026

Risk Level: SAFEPROMPT_INJECTIONNO_CODE
Full Analysis
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it ingests untrusted data from the user conversation and writes it to a local markdown file.
  • Ingestion points: The skill instructs the agent to "Scan the conversation for domain-relevant nouns, verbs, and concepts" in SKILL.md.
  • Boundary markers: There are no explicit boundary markers or instructions to treat the conversation content strictly as data rather than instructions.
  • Capability inventory: The skill has file-write capabilities, specifically targeting the creation and updating of UBIQUITOUS_LANGUAGE.md in the working directory.
  • Sanitization: The instructions lack any requirement for sanitization, filtering, or escaping of terms extracted from the conversation before they are written to the output file.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 12, 2026, 04:29 PM
Security Audit — agent-trust-hub — ubiquitous-language