In-depth architectural comparison of the Facturas Inbox MCP and Docweave MCP MCP servers. Compare execution transports, security boundaries, tool capabilities, quality scores, and ready-to-paste client installation snippets for Claude, Cursor, Windsurf, and VS Code.
At a Glance & Executive Verdict
Facturas Inbox MCP
Developer Tools · Local stdio
Quality: 56/100 (Good) | Auth: No auth required
Docweave MCP
Developer Tools · Local stdio
Quality: 57/100 (Good) | Auth: API Key required
Verdict Summary: Choose Facturas Inbox MCP if you need specialized Developer Tools tools running via a local process. Choose Docweave MCP if your workspace requires Developer Tools integration with local subprocess execution. Both servers can be configured concurrently in your client's mcpServers manifest.
Which MCP Server Should You Choose?
F
Choose Facturas Inbox MCP when:
You need dedicated capabilities in the Developer Tools domain.
You prefer local stdio subprocess transport architecture.
Your security boundary fits: No auth required (Free / Open Source).
You have access to required keys: FACTURAS_INBOX_DIRS.
Spanish invoice PDFs to validated data: NIF/IVA/IRPF extraction, arithmetic checks, IVA per quarter.
Generate and read PDFs for AI agents: a generatepdf tool (HTML, a URL, or a template + JSON → PDF) and a readpdf tool (a PDF → text/markdown). Run locally with npx @docweave/mcp or use the hosted streamable-HTTP endpoint. Chromium-rendered, priced per document.
Facturas Inbox MCP is categorized under Developer Tools and uses a local stdio subprocess. In contrast, Docweave MCP belongs to Developer Tools using local stdio subprocess. Select Facturas Inbox MCP when you need capabilities focused on developer tools and Docweave MCP when you require tools for developer tools.
Write the scan to a new `.csv` file (`;`-separated, decimal comma, opens directly in Spanish Excel). Never overwrites
Docweave MCP Tools (2)
generate_pdf
Generate a PDF from raw HTML, a public URL, or a template + JSON data. Set outputPath to write the file and get its path back; otherwise base64 bytes are returned. The canonical way for an AI agent to turn content into a shareable, correctly-formatted PDF.
read_pdf
Read a PDF and return its text as markdown (or plain text). Accepts a public URL, base64 bytes, or a local file path. Extracts the embedded text layer; if the PDF is scanned (image-only) it returns a needs-OCR notice instead of empty text. The canonical way for an AI agent to ingest a document's contents.