DWG MCP Server
An MCP server for read-only inspection of AutoCAD DWG files, enabling AI agents to open drawings, query objects by handle or filter, and explore properties and references.
README
DWG MCP Server
DWG MCP Server is an MCP server for inspecting AutoCAD DWG files with AI agents and assistants. It provides read-only access to drawing contents as queryable objects. Agents can open a DWG, inspect available types, fetch objects by handle, and query properties, scopes, and references.
Quick Start
Use DWG MCP Server from the MCP client of your choice.
The npm package launches the published Docker image.
The Docker launcher mounts $HOME by default. For drawings outside $HOME, set
DWG_MCP_DOCKER_MOUNTS.
If your MCP client, AI agent, or test harness does not support MCP roots, set
DWG_MCP_ALLOWED_ROOTS to the folders that should be accessible.
Codex
codex mcp add dwg-mcp \
--env DWG_MCP_ALLOWED_ROOTS="$HOME" \
-- npx -y @dmytro-prototypes/dwg-mcp-server
Codex does not currently provide MCP client roots to stdio MCP servers. Configure explicit allowed roots for the folders that contain your drawings.
Claude
claude mcp add --scope user --transport stdio dwg-mcp -- npx -y @dmytro-prototypes/dwg-mcp-server
Claude Code provides MCP roots, so no extra path variable is usually needed.
Cursor
{
"mcpServers": {
"dwg-mcp": {
"command": "npx",
"args": ["-y", "@dmytro-prototypes/dwg-mcp-server"]
}
}
}
Exposed Tools
| Tool | Purpose |
|---|---|
dwg.list_roots |
List folders available for DWG access. |
dwg.open_file |
Open a DWG from an available folder and return a documentId. |
dwg.close_file |
Close an opened document and release its worker process. |
dwg.list_types |
List the globally supported DWG types known to the backend. |
dwg.list_file_types |
List only the types that are present in a specific opened DWG. |
dwg.describe_type |
Describe a supported type, including its properties and default projection. |
dwg.get_objects |
Fetch specific objects by handle, preserving the requested order and reporting missing handles. |
dwg.query_objects |
Query objects with filters, scopes, relation traversal, sorting, projection, and pagination. |
A typical flow is:
- Discover available folders with
dwg.list_roots. - Open a file from one of those folders with
dwg.open_file. - Inspect supported or file-local types with
dwg.list_types,dwg.list_file_types, ordwg.describe_type. - Fetch known handles with
dwg.get_objectsor search the drawing withdwg.query_objects. - Close the session with
dwg.close_file.
Architecture
Runtime model
In the packaged deployment, DWG MCP Server is a stdio MCP server implemented in Python and typically run inside a Linux container. The Python host exposes the MCP tools, validates file access, and manages document sessions.
Each dwg.open_file call starts a dedicated Rust dwg-worker process for that DWG and returns a host-side documentId.
All later file-scoped calls use that document id.
dwg.close_file terminates the worker for that session.
Worker and query model
The Rust worker speaks newline-delimited JSON over stdin and stdout. When it opens a DWG through LibreDWG, it first builds an in-memory indexed document. That upfront indexing step is central to the design: the server pays the cost once when the file is opened, then answers later requests against the index instead of rescanning the DWG each time.
The indexed model stores object handles, kinds, type names, generic types, summary and full properties, and derived block, layout, and space membership. It also stores supported type metadata such as aliases, default projections, and property definitions.
When you request full object records, responses also include that derived membership under extendedData, including container block, layout, and model or paper space when known.
dwg.get_objects is direct lookup by handle.
dwg.query_objects runs over indices for handle, type, generic type, kind, exact property values, block, layout, and space, then applies filters, scopes, relation traversal, sorting, projection, and pagination.
This is what makes queries over blocks, layers, layouts, references, and related objects practical on an opened drawing.
Access and packaging
DWG files must be opened from roots listed by dwg.list_roots. The server first
asks the MCP client for roots. If the client, AI agent, or test harness does not
support MCP roots, configure explicit allowed roots:
python3 -m dwg_mcp_server --allowed-root "$HOME/Downloads"
or:
DWG_MCP_ALLOWED_ROOTS="$HOME/Downloads;$HOME/Documents/dwg" \
python3 -m dwg_mcp_server
DWG_MCP_ALLOWED_ROOTS is an authorization fallback for clients without roots.
It should be a semicolon-separated list of absolute directories.
Docker is separate from MCP roots: roots authorize access, mounts make host
folders visible inside the container. The npm Docker launcher mounts $HOME by
default. For drawings outside $HOME, set DWG_MCP_DOCKER_MOUNTS.
For clients without MCP roots, DWG_MCP_ALLOWED_ROOTS is also used as the Docker
mount list unless DWG_MCP_DOCKER_MOUNTS is set.
The Docker image itself unpacks a prebuilt LibreDWG bundle, builds the Rust worker against it, and copies only the static-linked worker plus schema files into the final Python runtime image.
Build and Test From Source
Local source builds use the vendored third_party/libredwg submodule by default.
Prerequisites
- Rust toolchain
- Python 3.11 or newer
- Docker, if you want to build or run the container image
- autotools for local LibreDWG builds on macOS or Linux (
autoreconf,aclocal,automake,autoconf,make)
Bootstrap
git submodule update --init --recursive
bash scripts/build-libredwg.sh
Prebuilt LibreDWG bundles
CI and Docker builds use a Git LFS LibreDWG bundle instead of rebuilding LibreDWG on every run. The current CI bundle target is:
prebuilt/libredwg/x86_64-unknown-linux-gnu.tar.gz
Refresh it only when the LibreDWG submodule or native build inputs change:
bash scripts/build-linux-libredwg-prebuilt.sh x86_64-unknown-linux-gnu
For local macOS debugging, keep using bash scripts/build-libredwg.sh and the
local Rust build; the Linux prebuilt archive is for CI and Docker.
Build and test
cargo test --workspace
bash scripts/run-e2e-tests.sh
Run the MCP host locally
The Python host looks for dwg-worker under target/release or target/debug.
If you want a release build explicitly:
cargo build -p dwg-worker --release
Then run the MCP host:
PYTHONPATH=server/src python3 -m dwg_mcp_server
If the worker binary lives somewhere else, set DWG_WORKER_BIN to that executable.
Build and run with Docker
Build the image:
bash scripts/build-docker-mcp-server.sh
Run the server and expose specific host folders read-only:
DWG_MCP_DOCKER_MOUNTS="$HOME/Documents;$HOME/Desktop/dwg" \
bash scripts/run-docker-mcp-server.sh
Use this when your client supports MCP roots but the container still needs those host root paths mounted read-only.
For clients without MCP roots, set DWG_MCP_ALLOWED_ROOTS instead:
DWG_MCP_ALLOWED_ROOTS="$HOME/Documents;$HOME/Desktop/dwg" \
bash scripts/run-docker-mcp-server.sh
Use this when your AI agent or harness does not provide MCP roots. The server
will allow those directories, and the Docker launcher will mount the same
directories read-only unless DWG_MCP_DOCKER_MOUNTS is set.
By default, the Docker launcher mounts DWG_MCP_ALLOWED_ROOTS, then
~/Documents when no roots are configured. Access is still authorized by MCP
client roots or explicit allowed roots; Docker mounts only make those paths
visible inside the container.
Clean rebuild
Remove local build artifacts, Python caches, and the local Docker image:
bash scripts/clean-build-artifacts.sh
To also wipe the host LibreDWG build under third_party/libredwg:
bash scripts/clean-build-artifacts.sh --with-libredwg
Official MCP Registry
Registry metadata lives in server.json under the GitHub-authenticated name
io.github.dimitrovakulenko/dwg-mcp-server.
Before publishing a new registry version:
- Publish the matching npm package version from
npm/. - Confirm
npm/package.jsoncontains the samemcpNameasserver.json. - Run the manual
Publish MCP RegistryGitHub Actions workflow, or run:
mcp-publisher login github
mcp-publisher publish
The registry validates the published npm package, so server.json must point to
an npm version that already exists on the public npm registry.
License
This project is licensed under the GNU General Public License v3.0.
See LICENSE for the full license text.
推荐服务器
Baidu Map
百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。
Playwright MCP Server
一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。
Magic Component Platform (MCP)
一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。
Audiense Insights MCP Server
通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。
VeyraX
一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。
graphlit-mcp-server
模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。
Kagi MCP Server
一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。
e2b-mcp-server
使用 MCP 通过 e2b 运行代码。
Neon MCP Server
用于与 Neon 管理 API 和数据库交互的 MCP 服务器
Exa MCP Server
模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。