skills/etalab-ia/dragster/rag-parse/Gen Agent Trust Hub

rag-parse

Pass

Audited by Gen Agent Trust Hub on Mar 31, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill requires the installation of the @llamaindex/liteparse global package from the npm registry. This is a recognized package from LlamaIndex, a well-known organization in the AI ecosystem.
  • [EXTERNAL_DOWNLOADS]: Recommends installing standard system utilities such as LibreOffice and ImageMagick via official package managers (Homebrew, apt) to support Office document and image conversion.
  • [COMMAND_EXECUTION]: The skill generates shell commands using the lit CLI tool to parse, batch process, and screenshot documents. This involves executing subprocesses which is the primary intended functionality of the tool.
  • [COMMAND_EXECUTION]: The skill processes unstructured files (PDF, DOCX, XLSX, etc.) which presents a surface for indirect prompt injection where malicious instructions could be embedded in the parsed documents.
  • Ingestion points: SKILL.md (via lit parse and lit batch-parse commands)
  • Boundary markers: Absent; the skill does not explicitly provide delimiters or warnings to ignore instructions found within the document content.
  • Capability inventory: SKILL.md (executes shell commands, writes output files to the filesystem)
  • Sanitization: Absent; the extracted text is intended to be used directly by the agent.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 31, 2026, 01:55 PM