Skip to content
flinkeyflinkey for Developers

AI Tools

flinkey provides dedicated tooling for AI-assisted development. Whether you use Cursor, Claude Desktop, Windsurf, Copilot or another AI coding agent — these resources help your agent understand the flinkey platform and generate correct integration code.

MCP Server

Section titled “MCP Server”

The fastest way for an AI agent to interact with flinkey. The MCP Server exposes all API operations as callable tools — your agent can list cars, create users, manage assignments and query events directly.

{
  "mcpServers": {
    "flinkey": {
      "url": "https://flinkey.mcp.witte.digital",
      "headers": {
        "X-Flinkey-API-Key": "{{flinkeyApiKey}}",
        "X-Flinkey-Username": "{{FLINKEY_API_MANAGER_USERNAME}}",
        "X-Flinkey-Password": "{{FLINKEY_API_MANAGER_PASSWORD}}",
        "X-Flinkey-Env": "uat"
      }
    }
  }
}

Add this to your agent’s MCP configuration with your UAT credentials. The agent can then call flinkey tools directly during development.

Full MCP documentation →

Agent Skill

Section titled “Agent Skill”

A ready-to-use skill file that gives your agent deep knowledge about the flinkey integration model, credential rules, API patterns and implementation recipes — without calling an external service.

Installation: Place the skill files in .cursor/skills/flinkey-integration/ (or your agent’s equivalent skill directory).

Download & details →

LLM Context Files

Section titled “LLM Context Files”

Machine-readable context files for LLM-based workflows:

llms.txt

Index file with key page links for LLM crawlers. /llms.txt

llms-full.txt

Complete integration context as a single text file. /llms-full.txt

Which tool to use?

Section titled “Which tool to use?”
Scenario Recommended
Explore the API interactively from your IDE MCP Server
Generate integration code with full context Agent Skill
Feed context to an external LLM (ChatGPT, Claude) /llms-full.txt
Understand patterns and recipes Implementation Recipes