iso-13485-certification
Pass
Audited by Gen Agent Trust Hub on Oct 12, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONMETADATA_POISONING
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill provides instructions to process a directory of user-supplied documentation which could contain malicious instructions. * Ingestion points: The skill collects a directory of documents in various formats (.txt, .md, .doc, .docx, .pdf) for analysis in SKILL.md. * Boundary markers: There are no instructions or markers provided to the agent to distinguish between its own instructions and content found within the documents. * Capability inventory: The skill invokes a local Python script (scripts/gap_analyzer.py) to perform analysis on the provided documents. * Sanitization: The skill does not mention any sanitization or validation of the untrusted document content.
- [METADATA_POISONING]: The skill metadata lists the author as K-Dense Inc., whereas the provided context identifies the author as wu-yc. This discrepancy could lead to misidentification of the skill's origin.
Audit Metadata