browser-search-and-handoff
Pass
Audited by Gen Agent Trust Hub on May 28, 2026
Risk Level: SAFE
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process data from external websites (portals and marketplaces), which represents a potential surface for indirect prompt injection where malicious content on a webpage could attempt to influence the agent's behavior.
- Ingestion points: Data extracted from listing pages, profile pages, and application forms on third-party domains (documented in SKILL.md, Step 3 and 5).
- Boundary markers: The skill includes a robust 'automation boundary' (SKILL.md, Step 6) that mandates human intervention for high-risk steps like CAPTCHAs, payments, and legal acceptances.
- Capability inventory: The skill utilizes an external browsing tool to search the web and has the ability to write collected data to local CSV files or Google Sheets.
- Sanitization: No specific content sanitization is described, but the 'handoff' mechanism ensures the user reviews the collected results before taking further actions.
- [DATA_EXPOSURE]: The skill can optionally export tracked data to Google Sheets when requested by the user. This involves standard network operations to a well-known service and is governed by the user's explicit request for collaborative editing.
Audit Metadata