MCP Server

Connect Claude Code, Claude Desktop, Cursor, Codex, or any MCP client directly to Leonar using the Model Context Protocol.

Leonar exposes a native Model Context Protocol (MCP) server that lets AI assistants call Leonar tools and read contextual resources directly.

AI Assistant  →  MCP Protocol  →  Leonar MCP Server  →  Leonar CRM data

Quick start

1. Create an API key

Go to Settings > API in your Leonar dashboard and create a key with the scopes your assistant needs (see Scopes).

2. Configure your MCP client

Add the Leonar MCP server to your client’s configuration. The easiest way is via our NPX package, which works with all MCP clients.

NPX package (recommended)

Install nothing — just add this to your config:

Claude Code (.mcp.json)

{
  "mcpServers": {
    "leonar": {
      "command": "npx",
      "args": ["-y", "@leonar/mcp"],
      "env": {
        "LEONAR_API_KEY": "leo_your_api_key"
      }
    }
  }
}

Requires Node.js 18+. The package (@leonar/mcp) is a lightweight bridge that forwards MCP messages to the Leonar server.

Direct HTTP

If your MCP client supports the url transport type, you can connect directly without any package:

https://app.leonar.app/api/mcp

Claude Code (.mcp.json)

{
  "mcpServers": {
    "leonar": {
      "type": "url",
      "url": "https://app.leonar.app/api/mcp",
      "headers": {
        "Authorization": "Bearer leo_your_api_key"
      }
    }
  }
}

3. Start using tools

Once connected, your AI assistant has access to the same safe workflow tools used by Leonar’s integrated AI agent, plus 4 contextual resources. Ask it things like:

• “Search for backend engineers in Paris”
• “Show me the context for the Senior PM project”
• “Find sourcing candidates for this project”

Authentication

The MCP server uses the same leo_ API keys as the REST API. Pass your key as a Bearer token in the Authorization header.

Authorization: Bearer leo_your_api_key

Each tool requires specific scopes. If the API key lacks a required scope, the tool returns an error. See the scopes reference for the full list.

Use the minimum scopes your assistant needs. For read-only exploration, start with contacts:read, projects:read, and companies:read.

Approval-gated mutations used by the in-app agent (for example project creation or launching sourcing directly into a pipeline) are intentionally not exposed over external MCP until a reviewed approval flow is available for third-party MCP clients.

Transport

The MCP server uses Streamable HTTP transport over a single endpoint (POST /api/mcp). It is stateless — no session management is required. All MCP clients that support the url transport type can connect directly.

Available resources

The server also exposes these read-only resources:

Resource URI	Description
leonar://workspace/context	Workspace metadata, billing state, and lightweight counters
leonar://accounts/connected	Active connected accounts available for sending and sourcing
leonar://team/members	Workspace members and roles
leonar://pipeline/templates	Pipeline templates with their stages

Available tools

Search & context

Tool	Scopes	Description
search_leads	contacts:read	Search leads/contacts by text, job title, company, and location.
get_project_context	projects:read, pipeline:read	Get project context including pipeline stages and candidate counts.
get_project_sourcing_context	projects:read, pipeline:read, sourcing:read	Get sourcing-specific project context, owners, and available sourcing sources.
search_candidates	projects:read, sourcing:read	Search sourcing candidates using the same engine as the integrated AI agent.
suggest_followups	contacts:read	Produce structured follow-up suggestions for selected contacts.

Memory

Tool	Scopes	Description
save_user_memory	agents:write	Save a memory item for the current API key owner.
update_user_memory	agents:write	Update a saved memory item.
delete_user_memory	agents:write	Soft-delete a saved memory item.

Tool parameters reference

search_leads

{
  query?: string
  jobTitle?: string
  company?: string
  location?: string
  limit?: number
}

get_project_context

{
  projectId: string
}

get_project_sourcing_context

{
  projectId?: string
  projectName?: string
}

search_candidates

{
  projectId?: string
  projectName?: string
  query?: string
  jobTitle?: string
  titles?: string[]
  location?: string
  locations?: string[]
  companies?: string[]
  skills?: string[]
  languages?: string[]
  includeOpenToWork?: boolean
  sourcePreference?: "auto" | "linkedin_recruiter" | "unified"
  excludeAlreadyProcessed?: boolean
  excludeInProject?: boolean
  page?: number
  pageSize?: number
  searchId?: string
}

suggest_followups

{
  contactIds: string[]
  context?: string
  tone?: "formal" | "casual" | "friendly"
}

save_user_memory

{
  content: string
  memoryType?: string
  key?: string
  projectId?: string
}

update_user_memory

{
  memoryId: string
  content?: string
  key?: string
  status?: string
}

delete_user_memory

{
  memoryId: string
}

Example use cases for recruiters

Pipeline review

“Show me the context for the Senior PM project and how many candidates are in each stage.”

The assistant calls get_project_context and summarizes the pipeline without opening the dashboard.

Candidate search

“Find backend candidates for our Paris project, excluding already processed people.”

Uses search_candidates with the same sourcing defaults as the integrated AI agent.

Outreach status

“Give me three follow-up angles for these contacts before my outreach block.”

Uses suggest_followups to return structured talking points your assistant can turn into drafts.

Project sourcing context

“What sourcing sources are available for the Acme frontend project?”

Uses get_project_sourcing_context to inspect owners, pipeline state, and available sourcing providers.

Memory assist

“Remember that this workspace prefers concise, founder-style outreach.”

Uses save_user_memory so future agent interactions can reuse durable context.

Rate limits

The MCP server shares the same rate limits as the REST API: 1000 requests per hour per API key. Each tool call counts as one request.

MCP vs REST API

	MCP Server	REST API
Best for	AI assistants (Claude, Cursor, Codex)	Custom integrations, scripts
Protocol	Model Context Protocol	HTTP REST
Auth	Same leo_ API keys	Same leo_ API keys
Scopes	Same scope system	Same scope system
Tools	8 workflow tools + 4 resources	60+ endpoints (full CRUD)

Use the MCP server when connecting an AI assistant. Use the REST API when building custom integrations or scripts that need full CRUD access to all resources.