Skills
skills/kjanat/paperless-mcp/paperless-ngx/Gen Agent Trust Hub

paperless-ngx

Pass

Audited by Gen Agent Trust Hub on Feb 27, 2026

Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [PROMPT_INJECTION]: The skill presents an indirect prompt injection surface by ingesting untrusted data from document contents and metadata.
  • Ingestion points: Document OCR text and metadata are retrieved via the get_document and search_documents tools.
  • Boundary markers: No delimiters or ignore instructions are provided to the agent to distinguish document content from instructions.
  • Capability inventory: High-impact tools are available, including bulk_edit_documents (with delete, merge, and rotate methods) and delete_tag.
  • Sanitization: Document OCR content is not sanitized or filtered before being presented to the agent.
  • [COMMAND_EXECUTION]: The skill includes utility scripts (scripts/encode-file.sh and scripts/test-connection.sh) that use system commands like curl and base64 to support legitimate API interactions and file preparation.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 27, 2026, 11:37 AM