# Research Search — how to use (mcp.ai)

Search scholarly articles and patents from your agent. Articles come with authors, publication, year and citation count; patents with number, assignee and date. Useful for literature review and for prior art search before filing. Accepts up to 100 queries in a single call. Hosted by the platform, no credentials, you pay per query.

## Option A — via MCP (recommended)
Remote MCP endpoint (HTTP, streamable): `https://api.mcp.ai/p_research_search?ms=1792059360000`
Add it as a custom/remote MCP connector in your client (Claude, Cursor, VS Code…), then authenticate when prompted. Once connected, ask the agent to use the server's tools (e.g. `research_search_papers`).

## Option B — via direct REST API
Base URL: `https://api.mcp.ai/api/research_search`
Auth: `Authorization: Bearer sk_live_…` — create a workspace API key at https://mcp.ai/settings/api-keys
Discover endpoints: `GET https://api.mcp.ai/api/research_search/_endpoints`

### Endpoints
- `POST https://api.mcp.ai/api/research_search/papers` — Busca artigos acadêmicos com título, autores, publicação, ano e número de citações. Aceita até 100 consultas por chamada.
  - body: { consultas: string[], pais?: string, idioma?: string, limite?: integer, periodo?: string, pagina?: integer }
- `POST https://api.mcp.ai/api/research_search/patents` — Busca patentes com título, número, depositante e data. Use para busca de anterioridade antes de depositar. Aceita até 100 consultas por chamada.
  - body: { consultas: string[], pais?: string, idioma?: string, limite?: integer, periodo?: string, pagina?: integer }

## Example prompts
- "Bring the most cited articles about open banking in Brazil"
- "Is there prior art for this mechanism? Run the prior art search"
- "Build a literature review on this topic with articles from the last 5 years"

## More
- Page: https://mcp.ai/research_search
- Agent spec (llms.txt): https://mcp.ai/research_search/llms.txt
- Postman collection: https://mcp.ai/research_search/postman.json