iso-13485-certification
Pass
Audited by Gen Agent Trust Hub on Aug 17, 2026
Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [PROMPT_INJECTION]: The skill presents a surface for indirect prompt injection by design, as its core workflow involves the agent reading and summarizing content from untrusted, user-provided documents.\n
- Ingestion points: The directory of documents specified by the user in
SKILL.mdand processed by thescripts/gap_analyzer.pyscript.\n - Boundary markers: Absent. The agent is not instructed to use specific delimiters or to disregard instructions found within the user documents, increasing the risk of the agent obeying malicious commands hidden in the QMS files.\n
- Capability inventory: The agent can execute local scripts via
uv, interact with the file system, and provide summaries of external data to the user.\n - Sanitization: The provided
gap_analyzer.pyscript performs simple keyword matching and does not include any logic to sanitize or escape potentially malicious prompts found in text files.\n- [COMMAND_EXECUTION]: The skill's workflow requires the execution of a Python script (scripts/gap_analyzer.py) using theuvtool. This script interacts with the local file system by recursively searching for and reading documents in a user-defined directory.
Audit Metadata