deep-research

Pass

Audited by Gen Agent Trust Hub on Mar 8, 2026

Risk Level: SAFE
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill leverages EXA web search tools to retrieve technical information and concepts. This is the core functionality of the skill and utilizes a well-known research service for information gathering.
  • [COMMAND_EXECUTION]: Executes a local git command to resolve the repository's top-level directory. This is used for project-scoped caching and is implemented safely using a fixed argument list in subprocess.run without shell invocation.
  • [SAFE]: Manages research results through local file operations in ~/.claude/plugins/research/ and the project's docs/research/ folder. All file paths are constructed using a slug normalization process that sanitizes input and prevents directory traversal vulnerabilities.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 8, 2026, 07:30 PM