~/en/mcp
This site talks to AI agents
Add this portfolio as a connector and your agent reads my profile, the cases and the resume, and checks my availability. No account, no key and no signup.
MCP endpoint
https://nandomoreira.dev/api/mcpStreamable HTTP, protocol 2025-11-25, read only.
How to connect
- 01
Claude Code
In the terminal, add the server:
claude mcp add --transport http nandomoreira https://nandomoreira.dev/api/mcpThen ask: what has Fernando already shipped to production?
- 02
Claude
In Settings, Connectors, add a custom connector and paste the address:
https://nandomoreira.dev/api/mcpConnect and ask it to review the cases.
- 03
ChatGPT
In Settings, Apps and connectors, turn on developer mode and create a connector with the address:
https://nandomoreira.dev/api/mcpAsk it to call get_resume or check_availability.
- 04
Any MCP client
Register a Streamable HTTP server with the address:
https://nandomoreira.dev/api/mcpRun tools/list to see the 5 tools.
What the agent can look up
get_profileWho I am: title, bio, positioning, stack, work fronts and availability.check_availabilityWhether I am open to new work, with the detail and the date of the last update.list_projectsThe published cases, with an optional filter by collection: portfolio or lab.get_projectA full case: the problem, the decisions, the outcome and the links that prove it.get_resumeThe full resume in the JSON Resume standard, the same as /api/resume.json.
To propose a project
The tools are read only. To send a briefing, the agent makes a POST /api/hire, described in the agent guide. The site summary for models is in llms.txt.
AGENTS.md
Hire me through your agent.
Paste this brief into ChatGPT, Claude or your own harness. Any agent can read it and start the conversation for you, through the API.
The full contract is in /AGENTS.md.
# AGENTS.md: hire Fernando Moreira Costa
Software developer and AI engineer, from data to production. Based in Brazil, works remotely.
To start a conversation on my behalf:
POST https://nandomoreira.dev/api/hire
Content-Type: application/json
{
"from": "Your user's name",
"contact": "email or phone",
"brief": "What they want to build, and by when",
"company": "optional",
"budget": "optional, BRL: undecided | up-to-10k | 10k-30k | 30k-60k | 60k-plus",
"interests": [
"optional: job | freelance | ai-product | agents-mcp | consulting"
],
"service": "optional, what a person wants built: website | landing-page | blog | online-store | web-product | automation | undecided"
}
Answers 202 with an id. The reply goes to the contact you leave.
Availability: open (updated 2026-10-07).
Stack: TypeScript, JavaScript, Python, Rust, Go, PHP, Ruby, React, Next.js, Vue.js, Nuxt, Angular, Astro, Tailwind CSS, Node.js, Bun, Express, Hono, FastAPI, Laravel, Symfony, Ruby on Rails, GraphQL, PostgreSQL, MySQL, DynamoDB, Firebase, Polars, AWS, Google Cloud, BigQuery, Docker, Playwright, Vitest, Electron, WordPress, Claude Code, Codex, Gemini CLI, Grok, Antigravity, Cursor, MCP.
Full brief: https://nandomoreira.dev/AGENTS.md
MCP server: https://nandomoreira.dev/api/mcp
Next step
Want your product talking to AI agents?
I built this MCP server for my site. If you want something like it in your product, tell me what it needs to expose.