retrieve_by_query
AI agents call retrieve_by_query to retrieve information from Tabular Document Retriever MCP without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool performs semantic search and data retrieval from a local ChromaDB vector store. It queries existing tabular data (CSV/Excel) transformed into embeddings but does not create, modify, delete, or execute operations. Retrieval of data with no side effects is classified as Read, with low severity since the blast radius of misuse is limited to unauthorized data access.
From the tool's definition Tool name is 'retrieve_by_query' and sibling tools are 'retrieve_batch' and 'retrieve_single', all of which are retrieval operations.
Attacks that exploit this kind of access
retrieve_by_query. It is categorised as a Read tool in the Tabular Document Retriever MCP MCP Server, which means it retrieves data without modifying state.
Register the Tabular Document Retriever MCP server in PolicyLayer and add a rule for retrieve_by_query: 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 Tabular Document Retriever MCP. Nothing to install.
retrieve_by_query 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 retrieve_by_query 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 retrieve_by_query. 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.
retrieve_by_query is provided by the Tabular Document Retriever MCP server (lbaret/chromadb-retriever-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 →