zenf-docs
Pass
Audited by Gen Agent Trust Hub on Nov 6, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONNO_CODE
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill establishes a workflow for the agent to ingest content from local files and store it in a centralized registry. This creates a surface for indirect prompt injection, as malicious instructions within processed documents could influence the agent during subsequent retrieval or search operations.\n
- Ingestion points: Document files at absolute or relative paths specified by the user or agent.\n
- Boundary markers: The protocol lacks instructions for using delimiters or protective instructions when processing document snapshots.\n
- Capability inventory: Uses MCP tools and HTTP endpoints for reading, writing, and searching registry data.\n
- Sanitization: No content validation or sanitization is specified for the document snapshots stored in the registry.\n- [NO_CODE]: This skill consists entirely of instructional Markdown content and configuration metadata; it does not include any accompanying scripts, executables, or configuration files that execute at runtime.
Audit Metadata