How to Use the Qdrant MCP in Mastra AI

Q: How do I register Qdrant tools inside a Mastra AI agent?

Instantiate the client using new MCPClient with your Vinkius URL. This connects your agent to the MCP Server, letting you call mcpClient.listTools() and spread the resulting array directly into your agent's tools configuration block.

Q: Can Mastra AI handle automatic retries if a Qdrant search fails?

Yes, Mastra's workflow engine wraps the search tool in a resilient execution loop. If the vector database is temporarily unreachable, the engine backs off and retries the query before failing the step.

Q: How do I set up manual verification for Qdrant deletions in Mastra AI?

Use Mastra's requireToolApproval option on the delete tool. When your agent decides a vector point needs to go, the workflow halts until an administrator approves the payload retrieved by getpoints.

Automate complex vector workflows and conditional Qdrant pipelines with Mastra AI.

See Vinkius in Action

Works with every AI agent you already use

…and any MCP-compatible client

MCP Servers — Included with Plan

Connect Qdrant MCP to Mastra AI

Create your Vinkius account to connect Qdrant to Mastra AI — we handle the hosting, security, and runtime updates so you don't have to. No server setup required.

GDPR Included with Plan

Setup Qdrant with Mastra AI

Ask AI about this MCP

ChatGPT

Claude

Perplexity

Conditional vector branching with Mastra AI

Mastra AI lets you build multi-step pipelines that react to your database state. This MCP Server lets your agent use `search` and find vectors, but if it finds zero relevant matches, the workflow automatically branches to an ingestion step. It doesn't just fail; it adapts dynamically. When an index operation fails, the framework triggers automatic retries with exponential backoff. This ensures that tools like `delete` execute reliably even during temporary network drops or cluster reindexing.

Human-in-the-loop point deletion

Deleting vector points is permanent and risky. This MCP Server lets your Mastra agent call `delete`, but you can configure the framework to require manual approval before execution. The workflow pauses, notifies you, and waits for a green light. Before pausing, the agent runs `get_points` to pull the exact payload of the target vector. It presents this raw data to the reviewer, ensuring no one accidentally wipes critical production coordinates.

Automated collection synchronization

Keep your vector indexes clean by setting up background workflows. Your agent uses `scroll` to fetch segments of points and inspects their payloads for outdated metadata. If a point is stale, the Mastra workflow flags it for cleanup. To prevent resource exhaustion, the agent checks `get_collection` metrics first. If the vector count is too high, the Mastra engine splits the scroll process into smaller, throttled batches to protect your database performance.

Setup guide

Set up Qdrant MCP in Mastra AI

Prerequisites

Node.js 18+ and a TypeScript project
@mastra/mcp + @mastra/core packages
Active Vinkius subscription with a valid endpoint token

1

Install dependencies
Run npm install @mastra/mcp @mastra/core plus your preferred model provider (e.g. @ai-sdk/openai).
2

Configure the MCPClient
Create an MCPClient with your Vinkius endpoint as a URL object. Replace [YOUR_TOKEN_HERE] with your token from cloud.vinkius.com.
3

Discover and inject tools
Call mcpClient.listTools() and spread the result into your agent's tools object. All Qdrant tools become native Mastra tools.
4

Run with any model
Swap openai("gpt-4o") for any AI SDK-compatible provider. Call agent.generate() and the agent routes tool calls through MCP automatically.

agent.ts

import { MCPClient } from "@mastra/mcp";
import { Agent } from "@mastra/core/agent";
import { openai } from "@ai-sdk/openai";

const mcpClient = new MCPClient({
  id: "qdrant-mcp-client",
  servers: {
    "qdrant-mcp": {
      url: new URL(
        "https://edge.vinkius.com/[YOUR_TOKEN_HERE]/mcp"
      ),
    },
  },
});

const agent = new Agent({
  name: "Qdrant Agent",
  model: openai("gpt-4o"),
  instructions: "You have access to Qdrant tools.",
  tools: {
    ...(await mcpClient.listTools()),
  },
});

const result = await agent.generate(
  "List recent Qdrant transactions"
);
console.log(result.text);

Get your connection token →

Independent Platform Disclaimer: Vinkius is an independent platform and is not affiliated with, endorsed by, sponsored by, verified by, or otherwise authorized by Qdrant. All third-party trademarks, logos, and brand names are the property of their respective owners. Their use on this website is strictly for informational purposes to identify service compatibility and interoperability.

Why Choose Vinkius

Vinkius connects your tools to AI with real-time monitoring and automatic cost savings — all from one dashboard.

Connect Qdrant now

Real-time monitoring

Live

visibility into every interaction

Connect your favorite tools to your AI and see exactly what's happening — every request, every response, in real time.

Built-in savings

60%

lower AI costs

Vinkius compresses data between your apps and your AI automatically. Lower bills every month — no configuration required.

Single dashboard

One

place for every integration

Every tool your AI connects to, managed from a single screen. One account, complete control.

Common questions about Qdrant MCP in Mastra AI

Instantiate the client using `new MCPClient` with your Vinkius URL. This connects your agent to the MCP Server, letting you call `mcpClient.listTools()` and spread the resulting array directly into your agent's tools configuration block.

Yes, Mastra's workflow engine wraps the `search` tool in a resilient execution loop. If the vector database is temporarily unreachable, the engine backs off and retries the query before failing the step.

Use Mastra's `requireToolApproval` option on the `delete` tool. When your agent decides a vector point needs to go, the workflow halts until an administrator approves the payload retrieved by `get_points`.

The server automatically negotiates between Streamable HTTP and Server-Sent Events (SSE). Mastra detects the transport format out of the box, so you don't need to write custom networking adapters.

Yes, your point IDs and JSON payload attributes are processed entirely in-memory. The MCP setup runs inside an isolated, zero-trust sandbox, meaning no vector coordinates are ever stored or logged outside your secure session.

Use it with your favorite AI tools

Connect this server to Cursor, Claude, VS Code, and more.

OpenAI Agents SDK sdk-python

Google ADK sdk-python

Pydantic AI sdk-python

Vercel AI SDK sdk-typescript