Skip to main content
Glama
drapefit

drapefit

README.md
# drapefit

Simulate how composite fabric drapes over a CAD surface, without leaving your tool list. drapefit is a local forwarding client for the hosted drapecut MCP service: it answers `tools/list` locally from the captured tool definitions and forwards every call to drapecut, returning the answers exactly as the service sends them back.

The workflow runs inside anonymous, ephemeral two-hour workspaces. You create a workspace, load a built-in sample or import a part you uploaded, inspect and select faces, then run single drapes, compare warp angles, build multi-piece layouts, and export the resulting flat pattern as DXF or tiled PDF. Nothing is simulated, cached or stored locally — the package is a pipe to the solver, not the solver itself.

All 16 captured tools are built; there are no omissions.

## The tools

- **`create_workspace`** — Create a private two-hour workspace and return its ID and CAD upload page. Start here. No parameters.
- **`list_uploads`** — List files uploaded through the workspace's upload page. `workspace_id` is required.
- **`compare_drapes`** — Compare up to 12 warp angles at one fixed seed and cloth configuration, ranked by coverage, lock exceedance and maximum shear. `workspace_id`, `mesh_id` and `angles` are required.
- **`create_layout`** — Start covering one surface with multiple pieces of the same ply/fabric, up to `max_pieces`. `workspace_id` and `mesh_id` are required.
- **`export_pattern`** — Export a built-in example as DXF or tiled 1:1 PDF and return a temporary download URL. `workspace_id` and `result_id` are required.
- **`get_job`** — Poll a STEP import or a multi-piece layout job. `workspace_id`, `kind` and `job_id` are required.
- **`import_part`** — Import a file already uploaded to the workspace. `workspace_id` and `upload_id` are required.
- **`list_fabrics`** — List fabric IDs, nominal locking angles, tow pitch, thickness and UD settings. No parameters.
- **`list_faces`** — Inspect STEP face IDs, areas, bounds, centroids, neighbours and planar mesh normals. `workspace_id` and `step_id` are required.
- **`list_samples`** — List built-in example parts that can be loaded without uploading a file. No parameters.
- **`load_sample`** — Register a built-in sample and return a mesh_id for draping. `workspace_id` and `name` are required.
- **`mesh_landmarks`** — Inspect seed landmarks in mm, paged within each category. `workspace_id` and `mesh_id` are required.
- **`run_drape`** — Drape one fabric at one warp angle and return statistics plus an exportable result_id. `workspace_id` and `mesh_id` are required.
- **`select_faces`** — Create a drapable mesh from explicitly chosen STEP face IDs. `workspace_id`, `step_id` and `face_ids` are required.
- **`server_status`** — Check the hosted solver and read connection/workspace limits. No parameters.
- **`suggest_seed`** — Suggest a starting point on a mesh. `workspace_id` and `mesh_id` are required.

## Install and run

As an MCP server via npm:

```bash
npx -y @reneturcios/drapefit mcp
```

MCP client configuration:

```json
{
  "mcpServers": {
    "drapefit": {
      "command": "npx",
      "args": ["-y", "@reneturcios/drapefit", "mcp"]
    }
  }
}
```

As a container:

```bash
docker run -i ghcr.io/cloudcorprecords/drapefit
docker run --rm ghcr.io/cloudcorprecords/drapefit server_status '{}'
```

## CLI

One binary serves both the command line and the MCP server. Each tool is a subcommand taking a single JSON string with that tool's arguments:

```bash
drapefit create_workspace '{}'
drapefit list_samples '{}'
drapefit run_drape '{"workspace_id":"<id>","mesh_id":"<mesh>"}'
drapefit --version
```

Exit codes: `0` when the forwarded call succeeds, `1` when the upstream result is an error, and `2` for usage problems such as an unknown subcommand or a malformed JSON argument.

## Configuration

- `DRAPECUT_MCP_URL` — Overrides the upstream endpoint. The default is the pinned hosted endpoint, `https://drapecut.com/mcp`. (The override variable is a reconstruction added by this rebuild.)

The upstream registry record declares no headers, so no authentication header is ever sent; the tool surface was captured keyless. Workspace IDs and upload/download links returned by the service grant access to a workspace — do not publish them.

Upstream failures — a JSON-RPC error from the service, a non-2xx response, an unreadable body, or a request that fails outright — are reported to the caller as an error result whose text starts with `drapefit error:`.

## How forwarding works

The first forwarded call opens a session: an `initialize` request (protocol version `2025-03-26`), followed by a `notifications/initialized` notification, then the `tools/call` itself. If the service answers the initialize request with an `Mcp-Session-Id` header, that value is carried on every later request, and the open session is reused for the rest of the process. Replies are accepted either as plain JSON or as server-sent events. Successful results pass through verbatim, including any `isError` flag and any extra fields the service includes.

## Development

```bash
npm install
npm run build   # tsc -> dist/ (dist/cli.js is the bin)
npm test        # tsx --test test/*.test.ts — stubbed upstream, no network
npm run dev     # run the CLI from source
```

The test suite checks the served tool list against the captured surface (all 16 tools, deep-equal, in order, with no titles invented and no `required` key added to the four parameterless tools), exercises the session handshake and its reuse, verifies that no `Authorization` header is ever sent, and verifies that arguments go upstream untouched and results come back untouched.

## License

MIT — see `LICENSE`. Copyright (c) 2026 Rene Turcios.