MCP server
Oro Intel exposes the same metered data layer as a streamable-HTTP MCP server, so an agent can call it as tools. Endpoint: https://api.oro-intel.com/mcp/. Connect by signing in with your Oro Intel account — no key to copy.
Keep the trailing slash
The MCP URL must end in a trailing slash: https://api.oro-intel.com/mcp/. Without it the transport won't connect.
How to connect
Claude (web & desktop)
Settings → Connectors → Add custom connector. Name it Oro Intel, paste https://api.oro-intel.com/mcp/, click Add, and sign in when prompted.
ChatGPT
Settings → Apps & Connectors → Advanced settings, enable Developer mode (Plus/Pro), then Create connector with the MCP server URL https://api.oro-intel.com/mcp/. Choose OAuth and sign in when prompted.
Gemini (via Gemini CLI)
Restart the CLI and sign in in the browser window that opens on first use.
Claude Code
Sign in in the browser window that opens. If no prompt appears, run /mcp inside Claude Code and pick oro-intel → Authenticate.
Codex
Or add it to ~/.codex/config.toml:
Cursor
Settings → MCP → Add new MCP server, or add this to .cursor/mcp.json:
Windsurf
Add to ~/.codeium/windsurf/mcp_config.json (headless key auth):
VS Code (Copilot agent mode)
Add to .vscode/mcp.json — VS Code prompts for the key and stores it securely:
Cline
MCP Servers → Remote Servers → Add, or use the generic config below with a bearer key.
Any other MCP client
Add under mcpServers and sign in when prompted.
Programmatic access
Headless agents and scripts can authenticate with an API key as a bearer header instead of signing in — see the API reference for keys and per-call credit costs.
Tools
The MCP tools mirror the API endpoints one-to-one and carry the same credit costs (see the API table): oro_search_notices, oro_search_tenders, oro_lookup_company, oro_company_profile (flagship), oro_lookup_buyer, oro_buyer_profile, oro_bulk_export, oro_get_balance and more. Each tool description states its cost so the agent knows before it calls.
Runnable agent examples (LangChain, LlamaIndex, Vercel AI SDK, PydanticAI and more) are on GitHub. New to the API? See the quickstart.