Reportflow Mcp

Why ReportFlow MCP

Several MCP servers exist for business-document generation. ReportFlow MCP differentiates itself on three axes.

1. Remote-ready, setup-free — A hosted endpoint at https://mcp.re-port-flow.com/mcp lets you connect from Claude.ai (web) without installing npm or Node.js. Unlike other Japanese 帳票 (chohyo) MCP servers that ship as stdio-only local processes, business users without a dev environment can start using it on day one.
2. No-code template design + template marketplace — Templates are designed in a browser-based GUI editor (Konva-based) with no code. A free template gallery covers invoices, quotes, receipts, delivery slips, reports and more, so AI agents can invoke ready-made templates from day one. You don't need to author JSON schemas to add templates.
3. OAuth 2.0 + Dynamic Client Registration — Authentication is OAuth 2.0 (Authorization Code + PKCE) with Dynamic Client Registration — no API key is ever handed to the AI client. The security posture matches Claude.ai's first-party Connectors.

What it does

• Generate PDFs from natural-language requests like "create an invoice for Acme Corp totalling $330"
• Expose your ReportFlow designs and parameter schemas as MCP Resources so the AI can attach them as context
• Bulk-generate many PDFs and download them as a single ZIP
• Save outputs to your AI client's workspace folder (stdio mode)

Setup

Remote (recommended — no Node.js required)

{
  "mcpServers": {
    "reportflow": {
      "type": "http",
      "url": "https://mcp.re-port-flow.com/mcp"
    }
  }
}

VS Code uses servers at the top level instead of mcpServers. In Claude.ai (web), open Settings →
Connectors → Add custom connector and paste the URL above.

Local (stdio) — requires Node.js 18+

{
  "mcpServers": {
    "reportflow": {
      "command": "npx",
      "args": ["-y", "reportflow-mcp"]
    }
  }
}

MCP capabilities

- Tools: authenticate, list_templates, get_design_parameters, generate_pdf_sync, generate_pdf_async,
 generate_pdfs_sync, generate_pdfs_async, download_file, download_zip, suggest_params
- Resources: reportflow://designs, reportflow://designs/{id}/parameters, reportflow://errors,
reportflow://server-info
- Prompts: /generate_pdf, /generate_pdfs, /reportflow_help
- OAuth 2.0 + PKCE + Dynamic Client Registration
- Tool Annotations (title / readOnlyHint / destructiveHint / idempotentHint / openWorldHint) on
every tool so MCP-aware clients render human-readable names and route the right approval prompts

Authentication

Remote: when the client connects for the first time, a browser tab opens for OAuth consent. Pick a
workspace, approve, done. Tokens are managed server-side.

Local: ask the AI to "authenticate with ReportFlow" — same OAuth flow. Tokens are stored in the OS
keychain (macOS Keychain / Windows Credential Manager / Linux libsecret), with a chmod-0600 file
fallback when libsecret is unavailable.

Requirements

- Remote: an MCP-capable AI client (Claude.ai, Claude Desktop, Cursor, VS Code) and a ReportFlow
account. That's it.
- Local: Node.js 18+ (auto-fetched by npx), a local environment with a browser for first-run auth,
and a ReportFlow account.

Links

- npm: https://www.npmjs.com/package/reportflow-mcp
- GitHub: https://github.com/re-port-flow/reportflow-mcp
- Official docs: https://doc.re-port-flow.com/docs/integrations/mcp
- MCP Registry: io.github.re-port-flow/reportflow-mcp
- Glama: https://glama.ai/mcp/servers/re-port-flow/reportflow-mcp

License: MIT