alterlab-iso13485
Pass
Audited by Gen Agent Trust Hub on Sep 25, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes user-provided documentation to identify gaps in ISO 13485 compliance, which presents a surface for indirect prompt injection.
- Ingestion points: The
scripts/gap_analyzer.pytool and the agent's core workflow ingest untrusted data from a user-specified directory containing QMS documents (e.g., .txt, .md, .doc). - Boundary markers: There are no explicit instructions or code-level mechanisms (like delimiters or warning prompts) to isolate the content of analyzed files from the agent's execution instructions.
- Capability inventory: The skill uses a broad set of tools, including
Read,Write,Edit, andBash(python:*), allowing for file system modifications and script execution. - Sanitization: There is no evidence of sanitization or filtering of the content within the processed documents before the agent interacts with them or their metadata.
Audit Metadata