API Tokens & MCP

MCP (Model Context Protocol) is an open standard that lets AI assistants like Claude Desktop interact directly with your Linc AI data. Through MCP, external clients can query your projects, sessions, process hierarchies, and knowledge base without leaving their own interface.

Linc AI exposes an MCP-compatible endpoint that accepts JSON-RPC 2.0 requests. Access is controlled through Personal Access Tokens (PATs) that you create from the Settings page.

Creating a Personal Access Token

Available Scopes

Each scope controls access to a specific resource type. Grant only the scopes your client needs.

Available Tools

Once connected, your MCP client can call the following tools to interact with Linc AI data:

Projects

• list_projects — List all projects in your organization
• get_project — Get details for a specific project

Sessions

• list_sessions — List interview sessions in a project
• get_session — Get session details including status and deliverables
• get_session_messages — Retrieve the interview conversation
• list_session_files — List files attached to a session
• create_session — Create a new interview session (write scope)
• list_session_assignments — View who is assigned to sessions
• create_session_assignments — Assign team members to sessions (write scope)

Processes

• get_process_hierarchy — View the full process hierarchy from Architect
• list_process_gaps — List identified process gaps
• get_process_hierarchy_node — Get details for a specific process node
• get_process_strategy_draft — View the AI-generated transformation strategy
• get_process_chain_draft — View process chain analysis
• list_process_redesigns — List AI redesign drafts

SOP & BPMN

• get_session_sop / export_session_sop - Read or export the SOP generated for a session
• get_session_bpmn / export_session_bpmn - Read or export the BPMN diagram for a session

Fit/Gap

• list_fitgap_runs / get_fitgap_run - List or view fit/gap analysis runs
• list_fitgap_results / get_fitgap_result - Read fit/gap classifications and evidence
• reconcile_process_with_sop - Trigger a targeted fit/gap reconciliation (write scope)

People & Interview Plans

• list_org_people / get_org_person - Read the organization people directory
• get_org_understanding - Read the captured understanding of the organization
• list_interview_plans / get_interview_plan - View interview plans
• send_interview_plan - Send an interview plan as Scout invitations (write scope)

Knowledge Base

• search_knowledge_base — Search across all uploaded files and documents

Using with Claude Desktop

To connect Claude Desktop to your Linc AI data, add the following server configuration to your Claude Desktop settings:

{
  "mcpServers": {
    "linc-ai": {
      "url": "https://dashboard.withlinc.com/api/mcp",
      "headers": {
        "Authorization": "Bearer pmat_YOUR_TOKEN_HERE"
      }
    }
  }
}

Rate Limits & Security

• Rate limiting is applied per-token and per-IP to prevent abuse.
• Audit logging records every MCP request including the tool called, outcome, and duration.
• Write operations require two-step confirmation — the client must first request the action, then confirm it.
• Token identifiers (prefix and last 4 characters) are shown in the Active Tokens list so you can track which token is being used.

Revoking Tokens

Revoked tokens are immediately invalidated. Any MCP client using that token will lose access.