Connect AI agents to workforce intelligence — knowledge graphs, skills analysis, trend signals, and career recommendations from 150M+ data points.
Headai MCP Server is a Model Context Protocol server that connects AI agents — Claude, Cursor, GitHub Copilot, or any MCP client — to Headai's Core Engine APIs. Build knowledge graphs from job ads, research articles, investment data, and more. Compare skills, detect trends, and get AI-powered career recommendations.
A Headai API key is required. Contact headai.com for access.
/mcp — MCP protocol (JSON-RPC 2.0, Streamable HTTP)/mcp — SSE stream for active sessions/mcp — Session termination/health — Server health check/docs — This documentation page/tools — Tool listing (JSON)/changelog — Release history| Tool | Category | Description |
|---|---|---|
headai_text_to_graph | Core | Convert text into a semantic knowledge graph |
headai_text_to_keywords | Core | Extract weighted keywords from text |
headai_build_knowledge_graph | Core | Build graphs from datasets (job ads, articles, curricula, investment, news) |
headai_scorecard | Analysis | Compare two knowledge graphs — gap analysis, coverage scoring |
headai_compass | Recommendations | AI-powered recommendations (jobs, courses, skills, career paths) |
headai_build_signals | Trends | Time series trend analysis — emerging, growing, declining skills |
headai_join_graphs | Transform | Merge multiple knowledge graphs |
headai_modify_graph | Transform | Filter/refine a graph by group, weight, or keywords |
headai_translate_graph | Transform | Translate a graph between languages |
headai_digital_twin | Storage | Store/retrieve competency profiles |
headai_fetch_graph | Utility | Retrieve a graph by URL |
headai_fetch_and_save | Utility | Fetch a graph and save to local file |
headai_describe_graph | Utility | Get human-readable description of a graph |
headai_estimate_size | Utility | Estimate result size before building |
headai_list_token_endpoints | Admin | List API endpoints for your key |
headai_list_token_data | Admin | List data built with your key |
headai_get_jobs_by_text | Jobs | Find matching job listings |
headai_autocomplete_job_title | Jobs | Autocomplete job titles |
headai_job_title_relations | Jobs | Get skills related to a job title |
headai_autocomplete_industry | Jobs | Autocomplete industries |
headai_industry_relations | Jobs | Get skills related to an industry |
headai_run_analyst | Reports | Run automated QA/analysis reports |
headai_run_composer | Reports | Generate strategic HTML documents |
Format: Authorization: Bearer your_headai_api_key
{
"mcpServers": {
"headai": {
"command": "node",
"args": ["/path/to/headai-mcp-server/dist/index.js"],
"env": {
"HEADAI_API_KEY": "your_api_key_here"
}
}
}
}MCP Server URL: https://mcp.headai.dev/mcp
Authorization: Bearer your_headai_api_key"What AI skills are Finnish employers looking for right now?"
Uses headai_build_knowledge_graph with dataset job_ads, country fi, and AI-related search terms. Returns ranked skills with weights and connections.
"Compare our data science curriculum against what employers need"
Build two snapshots, then use headai_scorecard to get coverage %, matched skills, gaps, and surplus.
"How have cybersecurity skills evolved from 2022 to 2025?"
Build yearly snapshots, then headai_build_signals classifies skills into 8 groups: Emerging, Growing, Constant, Declining, Disappearing.
"What does the future look like for autonomous vehicles?"
Combine job_ads (now) → investment_data (1-3yr) → doaj_articles (5-10yr) into signals.
"I know Python and SQL. What should I learn next for data engineering?"
Use headai_compass for personalized skill recommendations based on Zone of Proximal Development.
Bearer prefixdoaj_articles, investment_data, news datasets require search_year parameterheadai_estimate_size to check data availability before buildingheadai_estimate_size first© 2026 Headai Ltd. All rights reserved.