Search for similar vectors in a collection
AI agents call search_vectors to retrieve information from Qdrant MCP Server without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
search_vectors performs a query-only operation that retrieves matching vectors without creating, modifying, or deleting data. It has no side effects beyond returning results. This is a classic Read operation with minimal blast radius if misused by an AI agent—the worst outcome is exposure of vector data already stored in the database, not unauthorized modification or destruction.
From the tool's definition Tool description states 'Search for similar vectors in a collection' — a retrieval operation with no modification or deletion of data.
Attacks that exploit this kind of access
Search for similar vectors in a collection. It is categorised as a Read tool in the Qdrant MCP Server MCP Server, which means it retrieves data without modifying state.
Register the Qdrant MCP Server MCP server in PolicyLayer and add a rule for search_vectors: 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 Qdrant MCP Server. Nothing to install.
search_vectors 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 search_vectors 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 search_vectors. 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.
search_vectors is provided by the Qdrant MCP Server MCP server (jimmy974/qdrant-mcp-server). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Every MCP server has a record like this.
Type a name, get the same breakdown: verified identity, auth posture, risk grade, capabilities, recommended policy.
Teams ship this data inside their own products. See what a licence covers →