MCPLow riskUnclaimed

mcp-nfe-br

MCP server for Brazilian e-invoicing: NF-e and NFC-e (modelo 55/65), schema 4.00, SEFAZ.

cmendezscmendezs/mcp-nfe-brUpdated Sep 12, 2026

server.json

{
  "$schema": "https://static.modelcontextprotocol.io/schemas/2025-12-11/server.schema.json",
  "name": "io.github.cmendezs/mcp-nfe-br",
  "description": "MCP server for Brazilian e-invoicing: NF-e and NFC-e (modelo 55/65), schema 4.00, SEFAZ.",
  "repository": {
    "url": "https://github.com/cmendezs/mcp-nfe-br",
    "source": "github"
  },
  "version": "0.9.1",
  "packages": [
    {
      "registryType": "pypi",
      "identifier": "mcp-nfe-br",
      "version": "0.9.1",
      "transport": {
        "type": "stdio"
      },
      "environmentVariables": [
        {
          "description": "Master switch. Set to '1' to disable all write-side tools across NF-e, NFC-e, NFS-e, AND CT-e (br__submit_nfe, br__distribute_dfe, br__submit_nfse, br__cancel_nfse, br__submit_cte, br__cancel_cte, br__correct_cte). Read-only mode for safe exploration. Setting only this variable is sufficient to disable all CT-e writes too — BR_CTE_READ_ONLY is an additional, independent gate, not a required second step.",
          "format": "string",
          "name": "BR_READ_ONLY"
        },
        {
          "description": "Set to '1' to disable the three CT-e write-side tools (br__submit_cte, br__cancel_cte, br__correct_cte). Independent of BR_READ_ONLY — either variable being set to '1' is sufficient to block CT-e writes; use this one to gate CT-e alone while leaving NF-e/NFS-e writes enabled.",
          "format": "string",
          "name": "BR_CTE_READ_ONLY"
        },
        {
          "description": "Logging level: DEBUG, INFO (default), WARNING, or ERROR",
          "format": "string",
          "name": "LOG_LEVEL"
        }
      ]
    }
  ]
}

Permissions

DeclaredDetected
Runs code—python
Installs—pypi:mcp-nfe-br@0.9.1
Runs install scripts—None
NetworkNoneNone
Needs credentialsNoneNone
Outside the workspace—None
Agent tools—None

Checks

Low risk · Nothing worth a warning was found.

Not reviewed by a person · Checked by rules; the model review is not switched on yet.

Versions

  1. #10.9.1latestOct 7, 2026