Dcc Mcp Word
FreeNot checkedWord adapter for the DCC-MCP ecosystem — headless Word IR to DOCX compile with read-back verification over the dcc-mcp-office runtime
About
Word adapter for the DCC-MCP ecosystem — headless Word IR to DOCX compile with read-back verification over the dcc-mcp-office runtime
README
Word adapter for the DCC-MCP ecosystem — thin application layer over dcc-mcp-office.
Status: v0.1.0 — headless compile path live. Word IR → Open XML compile →
read-back verification runs end to end with no Word installation, and the
gate runs headless in CI. Field/TOC refresh, document reflow and
track-changes inspection are host_limited: declared, graded, and never
claimed as verified.
The old blocker note ("blocked on the
dcc-mcp-officeM1 release") is stale and removed.dcc-mcp-officeshipped M1 in v0.2.0 (2026-08-16) and is now at v0.2.3; the Word IR (crates/office-ir,pub mod word, proposal §13.3) and theWordBackend.csCOM backend have been present since then. The three placeholder repos stalled on an unreviewed blocker note, not on a technical wall.
Part of the DCC-MCP host matrix
dcc-mcp-word — Word adapter for DCC-MCP — headless Word IR to DOCX compile with read-back verification over the dcc-mcp-office runtime.
It is one of 47 host adapters in the DCC-MCP catalog. Every adapter speaks the same MCP protocol and builds on the same core runtime contract; each one exposes the tools its own host needs on top of that.
This block is generated from the catalog entry in dcc-mcp-catalog.yml. Re-run the generator after changing the catalog.
What v0.1.0 does
| Capability | Grade | Meaning |
|---|---|---|
word.document.compile |
verified |
Word IR → DOCX, headless, CI-green |
word.document.read_back |
verified |
Reopen the artifact; compare every paragraph and table cell |
word.paragraphs.write, word.lists.write, word.tables.write |
verified |
Structured content writes |
word.headers_footers.write, word.fields.write |
verified |
Per-section headers/footers; real Word field instructions |
word.fields.update, word.toc.rebuild |
host_limited |
Word computes field results — needs the desktop app |
word.document.reflow, word.track_changes.inspect |
host_limited |
Needs the Word COM object model |
word.content_controls.fill, word.figures.inline |
unimplemented |
Parsed and validated by the IR, not emitted by the writer |
Run dcc-mcp-word capabilities for the machine-readable report, or
dcc-mcp-word preflight for the per-machine self-check.
Install
pip install dcc-mcp-word # adapter only (stdlib + dcc-mcp-core)
pip install "dcc-mcp-word[headless]" # + python-docx, for the compile path
python-docx is an opt-in extra, matching how dcc-mcp-excel treats
openpyxl: importing the package never pulls it, so the adapter stays
installable inside DCCs that pin their own version. A test enforces this
(tests/test_runtime_purity.py).
Quick start
# 1. a Word IR envelope (office-ir/1.0, kind: word)
cat > call-sheet.json <<'JSON'
{
"schema_version": "office-ir/1.0",
"kind": "word",
"document_id": "fpt:call-sheet-2026-10-07",
"metadata": {"title": "Call sheet — 2026-10-07", "author": "pipeline"},
"document": {
"sections": [{"title": "Schedule", "paragraphs": [{"text": "Call 07:00"}]}],
"tables": [{"header": true, "rows": [["Shot", "Status"], ["sh010", "ip"]]}],
"footers": [{"section_index": 0, "text": "Page"}],
"fields": [{"kind": "toc"}, {"kind": "page"}]
}
}
JSON
# 2. compile + read back
python -m dcc_mcp_word.skills.word-document.scripts.generate_document \
--input call-sheet.json --out output
The script prints JSON: artifact path, read-back verdict (paragraphs and cells checked, mismatches), validation checks and warnings.
Input contract
The adapter mirrors the Rust dcc-mcp-office-ir word schema 1:1 — styles,
sections, paragraphs, lists, tables, figures, content controls,
headers/footers, fields and review policy. See
src/dcc_mcp_word/word_ir.py for the
annotated contract, or word-document/SKILL.md for the agent-facing guide.
The honest boundary
python-docx writes a valid DOCX but has no layout engine. That is why the
two halves of word.document.reflow are graded separately:
- Compiling (IR → DOCX) is headless and
verified. - Reflowing, re-paginating, and refreshing field results needs Word, and is
host_limited.
A TOC field is written as a real field instruction. Word fills in the
entries when the document is opened — verified on a Word 365 host: before
Fields.Update() the TOC shows its placeholder, after it shows
Schedule→1, Safety→2. See
examples/com-verification.json.
Test
pip install -e ".[dev]"
pytest
ruff check src tests
The headless gate must stay green without Word installed — that is the proof the adapter does not depend on a desktop Word installation.
Upstream
dcc-mcp-office— protocol, IR, C# COM runtime, jobs, security policydcc-mcp-core— gateway, skills runtime, sidecar lifecycle
License
MIT
Install Dcc Mcp Word in Claude Desktop, Claude Code & Cursor
unyly install dcc-mcp-wordInstalls into Claude Desktop, Claude Code, Cursor & VS Code — handles npx, uvx and build-from-source repos for you.
First time? Get the CLI: curl -fsSL https://unyly.org/install | sh
Or configure manually
Run in your terminal:
claude mcp add dcc-mcp-word -- uvx dcc-mcp-wordStep-by-step: how to install Dcc Mcp Word
FAQ
Is Dcc Mcp Word MCP free?
Yes, Dcc Mcp Word MCP is free — one-click install via Unyly at no cost.
Does Dcc Mcp Word need an API key?
No, Dcc Mcp Word runs without API keys or environment variables.
Is Dcc Mcp Word hosted or self-hosted?
Self-hosted: the server runs locally on your machine via the install command above.
How do I install Dcc Mcp Word in Claude Desktop, Claude Code or Cursor?
Open Dcc Mcp Word on unyly.org, pick your client tab (Claude Desktop, Claude Code, Cursor) and press Install — the config is generated automatically, no JSON editing.
Related MCPs
Opencode Omniroute Plugin
OpenCode plugin for the OmniRoute AI Gateway. Drives dynamic model discovery, /connect auth flow, and multi-instance OmniRoute providers via the official @openc
by GitHub ActionsRoblox Studio
Enables AI coding tools to control Roblox Studio for workspace exploration, instance manipulation, and script management. It provides tools for playtesting, sce
by paralovAntigravity Drawio
MCP server that enables AI assistants to create, parse, render, and validate Draw.io diagrams programmatically.
by S-SUJAN-SFetch
Web content fetching and conversion for efficient LLM usage.
AWS KB Retrieval
Retrieval from AWS Knowledge Base using Bedrock Agent Runtime.
by modelcontextprotocolSpring AI MCP Server
Provides auto-configuration for setting up an MCP server in Spring Boot applications.
llm-analysis-assistant
A very streamlined mcp client that supports calling and monitoring stdio/sse/streamableHttp, and can also view request responses through the /logs page. It also
by xuzexin-hzMCP-Agent
A simple, composable framework to build agents using Model Context Protocol by [LastMile AI](https://www.lastmileai.dev)
by lastmile-aiSpring AI MCP Client
Provides auto-configuration for MCP client functionality in Spring Boot applications.
mcp.natoma.ai
A Hosted MCP Platform to discover, install, manage and deploy MCP servers by [Natoma Labs](https://www.natoma.ai)
Compare Dcc Mcp Word with
Not sure what to pick?
Find your stack in 60 seconds
Author?
Embed badge for your README
Browse similar
All ai MCPs
