meta-docs

Fail

Audited by Gen Agent Trust Hub on Mar 12, 2026

Risk Level: HIGHEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill's core functionality involves processing documentation files from the docs/ directory, which introduces a surface for indirect prompt injection where instructions hidden within managed documents could potentially influence the agent's behavior. * Ingestion points: Document files are read from the local file system by doc_manager.py. * Boundary markers: The skill utilizes --- delimiters to separate YAML frontmatter from document bodies. * Capability inventory: The script performs file read/write operations and executes git config via subprocess for metadata management. * Sanitization: Employs the ruamel.yaml library for structured metadata parsing.
  • [EXTERNAL_DOWNLOADS]: The skill provides instructions to fetch and execute an installation script for the uv package manager from Astral's official domain.
  • [COMMAND_EXECUTION]: The doc_manager.py script invokes the git config command to automatically retrieve the local user's name and email for document history and metadata updates.
Recommendations
  • HIGH: Downloads and executes remote code from: https://astral.sh/uv/install.sh - DO NOT USE without thorough review
Audit Metadata
Risk Level
HIGH
Analyzed
Mar 12, 2026, 04:48 AM