For AI agents

Your agent can talk to us.

AI assistants now research vendors before people do. So we run an open MCP server: any compatible agent can browse our services, check whether we cover your legacy stack or ERP, run the readiness diagnostic and, if you ask, leave a message for the team. No login, no key.

Endpoint

https://rvffers.com/mcp

Protocol: MCP 2026-07-28 and 2025-11-25
Transport: Streamable HTTP, stateless
Authentication: None
Languages: pt, en, es

Try it live.

Pick a tool and see the JSON-RPC call an agent would make, with the real response from this server.

Tool

Arguments (JSON)

contact_rvffers is left out of the console because it sends a real message to the team.

Run a tool to see the response here.

Tools.

get_company_profile
Who we are: history since 2020, positioning and contact.
list_services
Catalog of services and products, with a summary and link for each.
get_service_details
One service in depth: approach, steps, coverage and link.
check_platform_support
Tells whether we cover a legacy platform or an ERP, and which services apply.
get_readiness_questions
The 11 questions of the agent readiness diagnostic, with their options.
score_readiness_assessment
Scores the diagnostic: level, areas and three recommendations.
contact_rvffers
Sends a message to the team, who reply by email. Only when you ask.

Connect your client.

Claude Code

In the terminal:

claude mcp add --transport http rvffers https://rvffers.com/mcp

Claude (web and desktop)

Settings › Connectors › Add custom connector, with this URL:

https://rvffers.com/mcp

Cursor

In ~/.cursor/mcp.json:

{
  "mcpServers": {
    "rvffers": {
      "url": "https://rvffers.com/mcp"
    }
  }
}

Plain HTTP

Any client that speaks JSON-RPC:

curl https://rvffers.com/mcp \
  -H "Content-Type: application/json" \
  -H "MCP-Protocol-Version: 2026-07-28" \
  -H "Mcp-Method: tools/call" \
  -H "Mcp-Name: list_services" \
  -d '{"jsonrpc":"2.0","id":1,"method":"tools/call","params":{"name":"list_services","arguments":{"language":"en"},"_meta":{"io.modelcontextprotocol/protocolVersion":"2026-07-28"}}}'

Privacy.

The lookup tools store nothing about who is asking. contact_rvffers should only be called when you ask, and stores only your name, email, company and message, plus the language and date, so we can reply.

Privacy policyPrefer plain text? The summary for language models is at /llms.txt