overdrive
Pass
Audited by Gen Agent Trust Hub on Sep 28, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill connects to a remote MCP server and REST API endpoints at
unbrowse.ai. These are vendor-owned resources essential for the skill's operation. - [INDIRECT_PROMPT_INJECTION]: The skill ingests third-party content from
overdrive.com, which could theoretically contain malicious instructions. - Ingestion points: Tools
overdrive_com__get_searchandoverdrive_com__read_pagefetch data from OverDrive. - Boundary markers: The instructions do not specify delimiters to separate fetched web content from user commands.
- Capability inventory: The agent can search, read, and perform browser actions on the target site.
- Sanitization: No explicit sanitization or filtering of web content is mentioned in the skill definition.
Audit Metadata