ubiquitous-language
Pass
Audited by Gen Agent Trust Hub on Jul 7, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill provides legitimate utility for extracting domain terminology. It uses standard workspace tools (Read, Write, Edit, Grep, Glob) to manage a documentation file (
UBIQUITOUS_LANGUAGE.md). - [PROMPT_INJECTION]: No evidence of behavior overrides, safety filter bypasses, or system prompt extraction was found. Instructions are strictly related to the stated task of glossary generation.
- [DATA_EXFILTRATION]: There are no network calls or attempts to access sensitive system files or credentials. The only file output is the local domain glossary.
- [REMOTE_CODE_EXECUTION]: The skill does not download external scripts, install unverified packages, or use dynamic execution patterns like eval or subprocess spawning.
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted input (the conversation history) to generate its output. While this represents a surface for indirect prompt injection, the risk is limited to the content of the generated documentation file and does not lead to code execution or data exfiltration. The skill correctly specifies that it should read existing files before updating, which is a good practice for maintaining data integrity.
Audit Metadata