A live endpoint your AI tools can query directly — work history, skills, role fit, preferences. Structured, honest and always current.
This is an MCP (Model Context Protocol) server exposing Matt's professional profile as structured, queryable data. Rather than parsing a CV, an AI assistant like Claude Desktop can call specific tools to answer whatever question you actually have.
Built because a CV is a lossy summary, LinkedIn is a keyword soup, and neither answers the real question a recruiter wants answered: is this person right for this role?
Paste the configuration below into your Claude Desktop configuration file and restart the app. Other MCP-compatible clients (Cursor, Continue, custom agents) use the same endpoint URL with their own config format.
{
"mcpServers": {
"matt-taylor": {
"url": "https://mcp.mattj-taylor.co.uk/mcp"
}
}
}
Tap the button above to copy the snippet to your clipboard.
Add to your claude_desktop_config.json and save.
The Matt Taylor tools appear in Claude's tool list automatically.
Natural language questions — Claude picks the right tool.
Once connected, ask Claude questions naturally. It calls the right tools in the background and returns structured, honest output. Here is a typical exchange:
Other useful prompts: “Summarise Matt's management experience,” “What industries is he looking to work in?” or “Draft an intro email for a staff engineer role at a devtools startup.”
Your AI picks the right tool automatically — you don't need to know which one to call. These are what's available behind the scenes:
get_profile_summary
get_work_history
get_role_details
get_skills
get_skill_evidence
assess_skill_match
list_projects
get_role_preferences