ubiquitous-language

Pass

Audited by Gen Agent Trust Hub on Aug 30, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted user input from conversation history and reads an existing local file (UBIQUITOUS_LANGUAGE.md) to generate content. It lacks explicit boundary markers or instructions to disregard potential commands embedded within the input data, which is a common characteristic of natural language processing skills. Ingestion points include conversation history and the file UBIQUITOUS_LANGUAGE.md. There are no specific boundary markers or sanitization steps mentioned for this process. The skill's capabilities are limited to writing results to a specific local file.
  • [SAFE]: The skill's primary function is restricted to reading and writing a specific domain-related documentation file. It does not utilize network tools, access sensitive system paths, or execute shell commands. The mention of shell commands in the example dialogue is purely illustrative of domain-specific terminology and does not constitute an instruction for the agent to execute code.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 30, 2026, 06:39 PM