Search in a memory file using hybrid, lexical, or vector search. Search modes: - hybrid (default): Combines lexical and vector search with RRF ranking - lex: Full-text lexical search only (Tantivy) - vec: Vector similarity search only (requires embeddings) Args: file: Absolute path to .mv2 file q...
AI agents call memvid_find to retrieve information from Memvid without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
memvid_find is a search/query tool that retrieves data from memory files without creating, modifying, deleting, or executing arbitrary code. It has no side effects and does not commit financial obligations. The blast radius of misuse is minimal—an agent could only exfiltrate existing data it has access to, which is a standard read operation risk.
From the tool's definition Tool performs "Search in a memory file" with modes like "hybrid", "lex", and "vec", and "Returns" search results. Arguments are query parameters (file path, search text, filters, limits) with no modification capability.
Attacks that exploit this kind of access
Search in a memory file using hybrid, lexical, or vector search. Search modes: - hybrid (default): Combines lexical and vector search with RRF ranking - lex: Full-text lexical search only (Tantivy) - vec: Vector similarity search only (requires embeddings) Args: file: Absolute path to .mv2 file query: Search query text mode: Search mode (hybrid, lex, vec) limit: Maximum results to return (default: 10) uri: Filter by exact URI match scope: Filter by URI prefix Returns: {. It is categorised as a Read tool in the Memvid MCP Server, which means it retrieves data without modifying state.
Register the Memvid MCP server in PolicyLayer and add a rule for memvid_find: 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 Memvid. Nothing to install.
memvid_find 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 memvid_find 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 memvid_find. 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.
memvid_find is provided by the Memvid MCP server (tapiocapioca/memvid-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 →