local-doc-to-markdown

Pass

Audited by Gen Agent Trust Hub on Sep 2, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill is designed to process content from untrusted external sources (local PDF, Office, and OpenDocument files) and convert them into markdown for the agent to process.
  • Ingestion points: Local document files processed via convert_document and convert_base64 tools.
  • Boundary markers: The instructions do not specify the use of delimiters or clear separation between document content and agent instructions.
  • Capability inventory: The skill tools allow for filesystem access (reading documents and writing output to specified paths).
  • Sanitization: There are no instructions for sanitizing or filtering the content extracted from the documents.
  • [EXTERNAL_DOWNLOADS]: The skill documentation references the use of npx @firecrawl/anydoc and recommends the installation of the mcp-server-anydoc MCP server from external registries.
  • [COMMAND_EXECUTION]: The skill explicitly mentions "shelling out to CLI converters" and using npx to execute document conversion tools, which involves interaction with the host system's shell environment.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 2, 2026, 02:52 AM
Security Audit — agent-trust-hub — local-doc-to-markdown