web-research
Pass
Audited by Gen Agent Trust Hub on Jul 31, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill uses
WebSearchandWebFetchtools as intended for its primary purpose of online research. All network operations are directed towards fetching content for the user's requested topic. - [SAFE]: The skill maintains transparency by printing reports to the chat and saving them to the local
~/research-reports/directory. This is standard behavior for a research agent. - [SAFE]: The practice of appending to
~/.claude/skills/web-research/LEARNINGS.mdis a self-improvement mechanism (feedback loop) and does not involve sensitive data exposure or persistence-based attacks. - [SAFE]: The skill uses structured templates and explicit citation requirements, which mitigates risks related to hallucination or misinformation in its output.
Audit Metadata