deep-research
Pass
Audited by Gen Agent Trust Hub on Mar 10, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill is a legitimate research tool that automates information gathering and documentation. No suspicious or malicious behaviors were detected during the analysis.
- [EXTERNAL_DOWNLOADS]: The skill employs WebSearch and WebFetch tools to gather information from the internet. This behavior is transparently described and essential for the skill's primary function of deep research.
- [COMMAND_EXECUTION]: The skill uses file system tools including Read, Write, Edit, Glob, and Grep to manage and update markdown playbooks. These operations are limited to the user's workspace and are driven by the research workflow.
- [PROMPT_INJECTION]: The skill ingests untrusted data from the web (indirect prompt injection surface). It mitigates risk by instructing sub-agents to follow strict formatting rules and requiring verifiable source URLs, focusing the AI on content extraction rather than command processing.
Audit Metadata