negative-support-mcp
Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@negative-support-mcp Generate supports for ~/models/part.stl"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
@spikeonstudio/negative-support-mcp
MCP server for negative.support — generate negative-space print supports (STL/OBJ/STEP) via the official negative-support package. Full API details: negative.support/docs.
Requirements
Node.js 18+
A negative.support token (
ns_live_…) from negative.support after sign-in
For a local clone, you can put NEGATIVE_SUPPORT_TOKEN=… in a .env file next to package.json (see .env.example). The server loads it via dotenv when the process working directory is that folder. Do not commit .env — it is listed in .gitignore.
Related MCP server: Pixal3D-MCP
MCP configuration
Use npx — no clone or local npm install required. Set NEGATIVE_SUPPORT_TOKEN, or omit it and call negative_support_activate once per session.
npx normally runs with another cwd, so prefer env in the MCP JSON for Cursor; use .env when you run npm run dev / npm start from this repo.
npm (default)
{
"mcpServers": {
"negative-support": {
"command": "npx",
"args": ["-y", "@spikeonstudio/negative-support-mcp"],
"env": {
"NEGATIVE_SUPPORT_TOKEN": "ns_live_..."
}
}
}
}From GitHub (same binary name as in package.json bin):
{
"mcpServers": {
"negative-support": {
"command": "npx",
"args": ["-y", "--package=github:spikeon/negative-support-mcp", "negative-support-mcp"],
"env": {
"NEGATIVE_SUPPORT_TOKEN": "ns_live_..."
}
}
}
}On Windows, if npx is missing, use "command": "npx.cmd" with the same args.
Tools
Tool | Description |
| Activate your license for this process. |
| Clear the local session (token not revoked on the server). |
| Build supports from a local STL/OBJ/STEP/STP; writes |
negative_support_generate: input_path (absolute, on the host running the MCP) is required. Optional: output_stl_path, output_3mf_path, format, margin, angle, min_volume, skip_merge — see negative.support/docs.
Resource: negative-support://docs — short in-server summary.
Timeouts
Generation can take several minutes on large meshes. Many MCP clients use a default tool timeout (often around 60s) unless the call stays “active.” This server emits MCP notifications/progress when the client sends a progressToken with the tool request (about every 15s plus throttled updates from the library), which usually refreshes that timeout.
If a host still cuts you off, use the negative-support CLI for huge models, or raise the tool-timeout in your client if it exposes one.
Development
git clone https://github.com/spikeon/negative-support-mcp.git
cd negative-support-mcp
npm install
npm run build # or: npm run devThe server uses stdio; it is started by the MCP host, not run interactively.
Publishing (maintainers)
Pushing to main does not publish. .github/workflows/publish-npm.yml runs npm publish --access public when a v* tag is pushed.
Add repository secret
NPM_TOKEN(npm access token with publish rights for@spikeonstudio/*).Bump
versioninpackage.json(each publish must be a new version).git tag vX.Y.Z && git push origin vX.Y.Z
prepublishOnly runs npm run build. dist/ is committed so GitHub-based npx installs work without a local TypeScript build.
License
MIT. Not affiliated with negative.support; integrates their published npm package and licensing.
Available Tools
3 toolsnegative_support_activateA
Activate a negative.support license for this MCP process. Same ns_live_* token as the CLI. Alternatively set env NEGATIVE_SUPPORT_TOKEN.
Inputs: token (required). Returns: ok, message
negative_support_deactivateA
Clear the active license in this process (does not revoke the token on the server).
Inputs: no named parameters. Returns: ok, message
negative_support_generateA
Generate support STL (and optionally 3MF with model + support pieces) from a local STL, OBJ, STEP, or STP file. Requires an activated license. Heavy meshes can take many minutes; MCP hosts that send a progressToken receive keepalive progress and avoid tool timeouts.
Inputs: input_path (required), angle, format, margin, +4 more. Returns: stats, progress, output_3mf_path, output_stl_path
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
TDQS
Scored across 3 tools
Each tool targets a distinct operation: activation, deactivation, and generation. There is no overlap or ambiguity between them.
All tool names follow the same pattern: 'negative_support_' prefix plus a clear verb (activate, deactivate, generate). The convention is consistent and predictable.
With 3 tools, the set is tightly scoped to the server's purpose of license management and STL generation. Each tool earns its place with no redundancy.
The core lifecycle is covered: activate a license, deactivate it, and generate output. No obvious missing operations for the stated domain.
Maintenance
Related MCP Connectors
MCP server for OpenAI API (chat completions, image generation, embeddings) via AceDataCloud
Free 3D conversion (FBX, GLB, OBJ, STL, PLY, DAE, USDZ, 3MF) as a remote MCP server. No account.
Official MCP server for Softhills 3D/2D Visualization SDKs, API documentation, and code examples.
MCP server for Flux AI image generation
Related MCP Servers
- AlicenseBqualityBmaintenanceEnables interaction with 3D printer management systems through an MCP server, supporting STL file manipulation, slicing, and control of printers like OctoPrint, Klipper, Duet, and more.305,162 npm266GPL 2.0
- FlicenseAqualityCmaintenanceStandalone MCP server for generating high-fidelity 3D assets from a single image using Pixal3D inference.121-
- AlicenseAqualityDmaintenanceMCP server for integrating with Rodin Gen-2 API to generate 3D models from text descriptions or images.6MIT
- AlicenseNot gradedqualityCmaintenanceMCP server for generating print-ready 3D models from text descriptions via a multi-agent pipeline, enabling AI assistants to create and optimize STL files.1MIT