yokingma/one-search-mcp

Description and use

**Brief Summary**
OneSearch MCP is a TypeScript‑based server that unifies web search, scraping, and data extraction across providers such as SearXNG, Tavily, DuckDuckGo, Bing, and more. It implements the Model Context Protocol (MCP), enabling AI assistants to invoke real‑world tools and retrieve up‑to‑date information through a single, standardized API.

**Value**
By abstracting the heterogeneity of public search APIs, OneSearch MCP lets developers plug any MCP‑compatible AI agent into a reliable, extensible search backend without writing provider‑specific adapters. This reduces integration effort, accelerates feature rollout for conversational agents, and creates a reusable “search as a service” layer that can be shared across multiple products or teams.

**Practical Adoption Path**

1. **Prototype** – Clone the repo, run the provided Docker compose or npm start script, and point your agent’s MCP client to the local endpoint.
2. **Configure Providers** – Add API keys and optional query parameters for the desired search backends in the config.yaml file; the server will automatically route requests.
3. **Integrate** – Use the generated OpenAPI spec or the built‑in SDK/CLI to call the /search endpoint from your AI workflow, handling the standardized JSON response.
4. **Scale** – Deploy the server to a container orchestration platform (K8s, ECS, etc.) behind a load balancer, enable caching or rate‑limit middleware, and monitor health via the built‑in Prometheus metrics.

**Production Readiness**
The project shows strong OSS maturity: 121 stars, recent commits (as of 2026‑06‑26), active issue handling, and a clear TypeScript codebase with an OpenAPI contract. Its modular architecture, Docker support, and existing integrations with major search engines make it suitable for pilot deployments and, after a brief security/license audit, for full‑scale production use.

Use in a workflow

yokingma/one-search-mcp helps connect AI assistants to real tools and data through a standard protocol.

• connect AI agents to tools
• ship Model Context Protocol servers
• standardize integrations

Limitations and review notes

High for an OSS candidate: recent activity, adoption, and ecosystem signals are strong enough for a serious pilot.