VectorMethods

Docs / MCP

Custom clients and HTTP transport

Configure custom MCP clients for VideoVector and deploy the MCP server in stdio or HTTP mode with the published environment variables.

api/mcp_controllers.pymcp-server/src/index.tsmcp-server/README.md

Search documentation

Search pages, API reference sections, and guide headings.

Summary

Custom clients can use the same stdio server config as Claude Desktop and Cursor or connect to a remote HTTP deployment when the MCP server must be network addressable.

Custom stdio configuration

The generic helper response uses this shape:

{
  "server": {
    "name": "videosearch",
    "command": "npx",
    "args": ["@videosearch/mcp-server"],
    "env": {
      "VIDEOSEARCH_API_KEY": "sk_live_..."
    }
  }
}

Stdio mode

stdio is the default transport mode.

npm install -g @videosearch/mcp-server
VIDEOSEARCH_API_KEY=sk_live_... videosearch-mcp

Environment variables:

  • VIDEOSEARCH_API_KEY
  • VIDEOSEARCH_BASE_URL
  • VIDEOSEARCH_TIMEOUT
  • VIDEOSEARCH_MAX_RETRIES

HTTP mode

Set MCP_TRANSPORT_MODE=http to expose the server over Streamable HTTP transport.

MCP_TRANSPORT_MODE=http \
VIDEOSEARCH_BASE_URL=https://playground-api-stg-udk7d32fva-uc.a.run.app/api/v2 \
node dist/index.js

HTTP mode adds:

  • PORT
  • MCP_HTTP_HOST
  • MCP_HTTP_ALLOWED_HOSTS
  • MCP_HTTP_ALLOWED_ORIGINS
  • MCP_HTTP_MAX_SESSIONS
  • MCP_HTTP_ENABLE_JSON_RESPONSE

HTTP endpoints

The server exposes:

  • GET /health
  • POST /mcp
  • GET /mcp
  • DELETE /mcp

Per-request auth can use either:

  • Authorization: Bearer sk_*
  • X-API-Key: sk_*

When to choose HTTP

Use HTTP transport when the MCP server:

  • runs remotely instead of inside a desktop client
  • needs origin or host allowlists
  • needs readiness probes
  • needs session management at the transport layer

Related documentation

The VideoVector MCP docs show how AI clients can browse indexes, run prompts, search media evidence, inspect workflow resources, and validate available tools.

The public MCP helper endpoints expose tool definitions, category groupings, platform-specific configs, and direct playground execution so you can validate integrations before deploying them.

This guide connects AI clients to VideoVector tools for media retrieval, prompt execution, workflow inspection, and playground validation.