MCP · Model Context Protocol
Connect your agent.
What is on this site is also available to your AI agent, through MCP. You can ask it to assess my profile, compare it with a job description or answer the questions of a first conversation.
Server address
https://portfolio-mcp.fabio-rafael7.workers.dev/How to connect
Claude Code
In the terminal:
claude mcp add --transport http portfolio https://portfolio-mcp.fabio-rafael7.workers.dev/
Claude (app and web)
In settings, under Connectors, add a custom connector and paste the server address.
VS Code
In the .vscode/mcp.json file:
{
"servers": {
"portfolio": { "type": "http", "url": "https://portfolio-mcp.fabio-rafael7.workers.dev/" }
}
}Cursor
In the .cursor/mcp.json file:
{
"mcpServers": {
"portfolio": { "url": "https://portfolio-mcp.fabio-rafael7.workers.dev/" }
}
}Clients change quickly. If any step no longer works, your client’s documentation has the current version.
What you can ask it
- Assess Rafael Pinto’s fit for this job description and tell me what is not evidenced: [paste the job description]
- What experience does Rafael Pinto have with Scrum and backlog management, and where is it shown?
- What should I ask Rafael Pinto in a first conversation?
Try it without an agent
It is the same server your agent uses. Connect, pick a tool and see the request and the response just as the agent sees them. A connection is only made when you press the button.
Good to know
Read-only and no login. The server returns what is on the site, without making anything up: your agent does the reasoning. The answers to the questions were written by me, not generated by AI. It has no database and does not store what you ask it.
How it is built
A stateless Cloudflare Worker that starts from the same content that generates this site. No database and no sessions.