managedcode-markitdown
Pass
Audited by Gen Agent Trust Hub on Sep 20, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process external document data (Office, rich-text) for conversion and downstream AI enrichment, creating a surface for indirect prompt injection.
- Ingestion points: The workflow involves processing input documents from potentially untrusted sources (SKILL.md).
- Boundary markers: The instructions do not specify the use of delimiters or boundary markers to isolate converted content from agent instructions.
- Capability inventory: The skill description indicates the agent may execute build, test, lint, or validation commands in the repository context (SKILL.md).
- Sanitization: No specific sanitization or content validation steps are prescribed for the converted Markdown before it is passed to downstream indexing or AI workflows.
Audit Metadata