videovector
Enables MCP clients to search, inspect, and operate VideoVector media intelligence workflows, including semantic/multimodal search, prompt management, and data operations via the public VideoVector API.
README
VideoVector MCP Server
Official Model Context Protocol (MCP) server for VectorMethods VideoVector.
This package lets MCP clients such as Claude Desktop, Cursor, Claude Code, and custom agent runtimes search, inspect, and operate VideoVector media intelligence workflows through the public VideoVector API.
Status
- Primary transport:
stdio - Self-hostable transport: Streamable HTTP at
/mcp - Runtime: Node.js 18+
- Package:
@vectormethods/videovector-mcp-server - Command:
videovector-mcp - Server name:
videovector
This repository is the public source of truth for VideoVector MCP server code, tool contracts, examples, and release metadata. Private VectorMethods backend deployment wiring, service accounts, project IDs, billing internals, and website source are intentionally not part of this repository.
Install
Use npx from your MCP client:
npx -y @vectormethods/videovector-mcp-server
Or install globally:
npm install -g @vectormethods/videovector-mcp-server
Stdio Usage
VIDEOVECTOR_API_KEY=<your-videovector-api-key> videovector-mcp
Environment variables:
| Variable | Required | Default | Description |
|---|---|---|---|
VIDEOVECTOR_API_KEY |
Yes for stdio | none | VideoVector API key. |
VIDEOVECTOR_BASE_URL |
No | https://api.vectormethods.com/api/v2 |
API base URL. |
VIDEOVECTOR_TIMEOUT |
No | 90000 |
Request timeout in milliseconds. |
VIDEOVECTOR_MAX_RETRIES |
No | 3 |
Retry count for retryable API failures. |
MCP_TRANSPORT_MODE |
No | stdio |
stdio or http. |
Client Examples
- Claude Desktop: examples/claude-desktop.json
- Cursor: examples/cursor.json
- Generic stdio: examples/custom-stdio.json
- Local Streamable HTTP: examples/streamable-http-local.json
Self-Hosted HTTP
HTTP mode is intended for self-hosted or private network deployments.
npm run build
MCP_TRANSPORT_MODE=http \
VIDEOVECTOR_BASE_URL=https://api.vectormethods.com/api/v2 \
node dist/index.js
Endpoints:
GET /healthPOST /mcpGET /mcpDELETE /mcp
HTTP requests must include either:
Authorization: Bearer <your-videovector-api-key>X-API-Key: <your-videovector-api-key>
HTTP hardening variables:
| Variable | Default | Description |
|---|---|---|
PORT |
8080 |
HTTP port. |
MCP_HTTP_HOST |
0.0.0.0 |
Bind host. |
MCP_HTTP_ALLOWED_HOSTS |
empty | Optional comma-separated host allowlist. |
MCP_HTTP_ALLOWED_ORIGINS |
empty | Optional comma-separated browser origin allowlist. |
MCP_HTTP_MAX_SESSIONS |
200 |
In-memory session cap. |
MCP_HTTP_ENABLE_JSON_RESPONSE |
false |
SDK JSON response mode. |
Do not advertise a public hosted remote MCP endpoint until OAuth and MCP protected-resource metadata are enabled for that deployment.
Tools
The server exposes tools for:
- semantic, image, multimodal, and structured metadata search
- index, video, segment, and prompt discovery
- prompt-run estimation, execution, status, results, retries, and cancellation
- prompt management and schema testing
- cloud connectors, import jobs, exports, and webhooks
The machine-readable tool contract is generated at artifacts/tool-contract.json. Private backend and website releases should vendor this artifact for MCP helper endpoints and documentation updates.
Development
npm ci
npm run verify
Useful scripts:
npm run typechecknpm run lintnpm run buildnpm testnpm run generate:contractnpm run check:contractnpm run check:examples
Use the MCP Inspector for local manual checks:
npx @modelcontextprotocol/inspector
For stdio, point the Inspector to npx -y @vectormethods/videovector-mcp-server and set VIDEOVECTOR_API_KEY.
Security
Never commit API keys, cloud credentials, connector credentials, webhook secrets, service-account JSON, or .env files.
See SECURITY.md for supported versions, disclosure instructions, and operational guidance.
推荐服务器
Baidu Map
百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。
Playwright MCP Server
一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。
Magic Component Platform (MCP)
一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。
Audiense Insights MCP Server
通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。
VeyraX
一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。
graphlit-mcp-server
模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。
Kagi MCP Server
一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。
e2b-mcp-server
使用 MCP 通过 e2b 运行代码。
Neon MCP Server
用于与 Neon 管理 API 和数据库交互的 MCP 服务器
Exa MCP Server
模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。