vector_store_search
AI agents call vector_store_search to retrieve information from Vending Machine without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
The name strongly suggests a read-only search operation against a vector store (knowledge base). With an empty description, confidence is moderately reduced, but the semantic meaning of 'search' in the context of vector stores—combined with sibling tools like 'query_knowledge' (also Read)—supports classification as Read. No evidence of data modification, deletion, or execution of arbitrary code.
From the tool's definition Tool name 'vector_store_search' indicates a search operation on a vector store, which typically retrieves or queries data without side effects.
Attacks that exploit this kind of access
vector_store_search. It is categorised as a Read tool in the Vending Machine MCP Server, which means it retrieves data without modifying state.
Register the Vending Machine MCP server in PolicyLayer and add a rule for vector_store_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 Vending Machine. Nothing to install.
vector_store_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 vector_store_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 vector_store_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.
vector_store_search is provided by the Vending Machine MCP server (yokiidesu/vending-machine-mcp). 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 →