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

parse

Pass

Audited by Gen Agent Trust Hub on Mar 31, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill instructs the user or agent to install the @llamaindex/liteparse package from the npm registry and system-level dependencies (LibreOffice, ImageMagick) using Homebrew or APT. These are standard dependencies for document processing from well-known sources.
  • [COMMAND_EXECUTION]: The skill relies on the lit CLI tool to perform document parsing, screenshots, and batch processing. It generates shell commands for file system operations and document conversion.
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because its primary function is to ingest and parse unstructured data from external files (PDF, DOCX, etc.). If these files contain hidden instructions, the agent might interpret them as commands.
  • Ingestion points: Reads arbitrary user-provided files via the lit parse command.
  • Boundary markers: None explicitly implemented to isolate parsed content from agent instructions.
  • Capability inventory: The skill has access to shell command execution through the lit utility.
  • Sanitization: No specific sanitization or filtering of the parsed text is described in the skill.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 31, 2026, 11:40 AM