pg_vector_search
Search for similar vectors. Requires: table, column, vector. Use select param to include identifying columns (e.g., select: [
This record as markdown: /tools/io-github-neverinfamous-postgres-mcp/pg-vector-search.md
What pg_vector_search does on PostgreSQL MCP Server
AI agents call pg_vector_search to retrieve information from PostgreSQL MCP Server without modifying anything. It is typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
Why pg_vector_search is rated Low
This tool performs a search operation to find similar vectors in a PostgreSQL database. The 'select' parameter allows specifying which columns to return, but the operation itself only queries and retrieves data. There is no indication of data modification, deletion, code execution, or financial impact. This is a Read operation with low severity—misuse would at worst retrieve unintended data.
From the tool's definition Tool name 'pg_vector_search' and description 'Search for similar vectors' with parameters for table, column, vector, and select indicate a query operation that retrieves data without modification or side effects.
Attacks that exploit this kind of access
The rule that runs pg_vector_search safely
PolicyLayer is an MCP gateway: it sits between your AI agents and PostgreSQL MCP Server, and checks every tool call against a rule you set before the call runs. Nothing changes on the server itself. For pg_vector_search, this is the rule to start with:
pg_vector_search is read-only, so it stays allowed. Everything else on the server is denied unless you say otherwise.
The button opens the PolicyLayer dashboard: create your workspace, connect PostgreSQL MCP Server, apply this rule, and every pg_vector_search call is checked against it from then on.
Questions about pg_vector_search
Search for similar vectors. Requires: table, column, vector. Use select param to include identifying columns (e.g., select: [. It is categorised as a Read tool in the PostgreSQL MCP Server MCP Server, which means it retrieves data without modifying state.
Register the PostgreSQL MCP Server MCP server in PolicyLayer and add a rule for pg_vector_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 PostgreSQL MCP Server. Nothing to install.
pg_vector_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 pg_vector_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 pg_vector_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.
pg_vector_search is provided by the PostgreSQL MCP Server MCP server (@neverinfamous/postgres-mcp). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
More on PostgreSQL MCP Server, and thousands of servers like it.
This server
Across the catalogue