convergentmethods avatar

Arezzo

convergentmethods/arezzo

Exposes three MCP tools for modifying Google Docs without silently corrupting them. The Docs API uses UTF-16 code units with cascading index shifts, so every insert or delete breaks subsequent indices in your batch. Arezzo compiles semantic operations (insert at heading, replace named range content, add table row) into correctly ordered batchUpdate requests with proper index arithmetic. Includes read_document to map headings and ranges, edit_document to execute compiled changes, and validate_operations for dry runs. Handles text formatting, tables, lists, images, headers, footers, and footnotes. Requires Google OAuth setup via arezzo init, then works through stdio transport in Claude Desktop, Cursor, or VS Code.

tools
3
auth
Required
transport
STDIO
env var
1
last updated
Apr 22, 2026

Tools

3 tools

Public tool metadata for what this MCP can expose to an agent.

read_documentReturns the document's structural map: headings with hierarchy, named ranges, tables, section boundaries.

Returns the document's structural map: headings with hierarchy, named ranges, tables, section boundaries.

No parameter schema in public metadata yet.

edit_documentCompiles operations into correct batchUpdate requests and executes them.

Compiles operations into correct batchUpdate requests and executes them.

No parameter schema in public metadata yet.

validate_operationsCompile-only dry run that returns the compiled requests for inspection without executing.

Compile-only dry run that returns the compiled requests for inspection without executing.

No parameter schema in public metadata yet.

Sponsors

Configuration

GOOGLE_APPLICATION_CREDENTIALS*

Path to the Google OAuth2 client credentials JSON (Desktop application type, Docs API enabled). Required — the Google Docs API does not support anonymous access. Run 'arezzo init' to walk through OAuth setup and write platform config.

Readme

Arezzo

Deterministic compiler for Google Docs API operations.

You cannot safely modify a Google Doc by constructing batchUpdate requests yourself. The API uses UTF-16 code units with cascading index shifts — insert 10 characters at position 50, and every subsequent index in your batch is now wrong. A single miscalculation silently corrupts the document with no error message.

Arezzo compiles semantic intent into a correct request sequence. Tell it what you want to do; it handles the index arithmetic.

For AI agents (MCP tools)

Arezzo exposes three tools via the Model Context Protocol:

read_document(document_id)
  → Returns the document's structural map: headings with hierarchy,
    named ranges, tables, section boundaries. Call this before editing
    so you know what addresses are available.

View the full README on GitHub

Details

Repository
github.com/convergentmethods/arezzo
Website
github.com/ConvergentMethods/arezzo
Category
AI & LLM Tools
Package
arezzo
Transport
STDIO
Auth
Required
Registry
MCP registry, active

Related AI & LLM Tools MCP servers

View all
1danecodes avatarBotcall Mcpdanecodes/botcall-mcpGive your AI agent a real phone number for SMS verification codes.0stars
2danielguru avatarRepomemorydanielguru/repomemoryPersistent memory for AI coding agents. Hybrid search, auto-session capture, context routing.0stars
3dbsectrainer avatarMcp Eval Runnerdbsectrainer/mcp-eval-runnerA standardized testing harness for MCP servers and agent workflows0stars
4vickhunter avatarAllagiovickhunter/allagio-pluginAgent-to-agent partnerships on ORP: find counterparties, negotiate deals, settle.0stars
5slipstack avatarSlipstack PDFdev.slipstack/pdfGenerate invoice and receipt PDFs from JSON. Agent-callable document API with a free demo quota.0stars
6sunrays avatarSunrays — Local Service Bookingdev.sunrays/bookingBook local tradespeople — plumber, electrician, HVAC, and 7 more — via your AI agent. All US.0stars