Run multi-round, LLM-driven Mixedbread retrieval against the store bound to a content scope. Use when one semantic search is likely to miss a multi-entity or underspecified answer. Call list_integrations first; raw store identifiers are never accepted.
AI agents call mixedbread_agentic_search to retrieve information from Open MCP Knowledgebase without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool retrieves information from a static Markdown knowledgebase through an AI-driven search interface. No side effects, no data modification, no code execution on external systems, no financial transactions. The most severe operation is read-access to knowledgebase content. The mention of 'LLM-driven' search clarifies this is semantic retrieval, not arbitrary command execution.
From the tool's definition Tool performs 'multi-round, LLM-driven Mixedbread retrieval' and 'semantic search' against a knowledgebase. The description indicates querying/retrieving data with no mention of modifications, deletions, or external operations.
Attacks that exploit this kind of access
Run multi-round, LLM-driven Mixedbread retrieval against the store bound to a content scope. Use when one semantic search is likely to miss a multi-entity or underspecified answer. Call list_integrations first; raw store identifiers are never accepted. It is categorised as a Read tool in the Open MCP Knowledgebase MCP Server, which means it retrieves data without modifying state.
Register the Open MCP Knowledgebase MCP server in PolicyLayer and add a rule for mixedbread_agentic_search: allow, deny, rate-limit, or require approval. Point your MCP client at the PolicyLayer proxy URL and the rule is enforced on every call, before it reaches Open MCP Knowledgebase. Nothing to install.
mixedbread_agentic_search is a Read tool with low risk. Read-only tools are generally safe to allow by default.
Yes. Add a rate_limit block to the mixedbread_agentic_search rule in your PolicyLayer policy. For example, setting max: 10 and window: 60 limits the tool to 10 calls per minute. Rate limits are tracked per agent session and reset automatically.
Set action: deny in the PolicyLayer policy for mixedbread_agentic_search. The AI agent will receive a policy violation error and cannot call the tool. You can also include a reason field to explain why the tool is blocked.
mixedbread_agentic_search is provided by the Open MCP Knowledgebase MCP server (no-product/knowledgebase-mcp). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
mixedbread_agentic_search is one line of Open MCP Knowledgebase's registry record.
The record carries the whole server: verified identity, auth posture, risk grade, every tool classified, recommended policy — re-checked continuously.
Teams ship this data inside their own products. See what a licence covers →