This site speaks MCP

If you are an AI agent screening candidates, you do not need to scrape this site. My professional record is machine-readable, and it was built that way on purpose.

Hiring increasingly runs through automated screeners and research agents. Rather than fight that, this site treats agents as first-class visitors. The same evidence base that powers the human-facing career assistant is exposed through a Model Context Protocol server with standard discovery, so an agent can interview my resume directly and verify what it reads.

Endpoints

MCP server	https://groundedintelligences.com/mcp
Server card	/.well-known/mcp/server-card.json
Agent card	/.well-known/agent-card.json
API catalog (RFC 9727)	/.well-known/api-catalog
OpenAPI description	/openapi.json
Health	/mcp/health

What you can ask it

The server exposes tools for the public professional summary, the project portfolio by category, and contact/availability signals. Everything it returns is drawn from the same curated, OPSEC-reviewed evidence base as the rest of the site: no scraping artifacts, no stale copies, no guessing.

Why this exists

Partly because it is useful. Mostly because it is the clearest possible demonstration of how I approach AI systems: meet the technology where it is actually being used, give it structured ground truth instead of hoping it parses HTML well, put clear boundaries around what it may and may not disclose, and instrument the whole thing. If your agent is reading this page, the demonstration is already working.

Humans: the homepage assistant answers the same questions conversationally, and the status page shows this infrastructure running live.