octen-web-search
Pass
Audited by Gen Agent Trust Hub on Sep 29, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill provides instructions for connecting to a remote Model Context Protocol (MCP) server at
https://mcp.octen.ai/mcpusing theclaude mcp addcommand. It also points to the vendor's official GitHub repositories (Octen-Team/octen-mcpandOcten-Team/octen-skills) for source code and documentation. These are recognized as legitimate vendor resources for the Octen search service. - [INDIRECT_PROMPT_INJECTION]: As the skill is designed to ingest and process data from the live web, it contains an attack surface for indirect prompt injection if the retrieved web pages contain malicious instructions.
- Ingestion points: Search results and full page content extracted from external URLs via the
searchandextracttools (SKILL.md). - Boundary markers: The instructions do not specify the use of delimiters or explicit "ignore" warnings to prevent the agent from following instructions embedded in retrieved data.
- Capability inventory: The skill enables several network-based data retrieval operations (
search,broad_search,extract,image_search,video_search). - Sanitization: No specific content sanitization or validation logic is described for the information retrieved from external websites.
Audit Metadata