본문으로 건너뛰기
cccskills
무료GitHub에 공개

bdb-rhino-mcp

Utilizes the Rhino MCP servers (RhinoMCP and golem-rhino-mcp) to create and manipulate 3D models and run Grasshopper graphs.

한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.

설치 방법 보기

포함된 파일(1)

  • SKILL.md7.4 KB

SKILL.md(원문)

설치하기 전에 에이전트에게 전달되는 지시 내용을 확인할 수 있습니다.

Rhino 3D MCP — Integration and AI Agent Guide

This skill file instructs AI agents on how to utilize the McNeel Rhino MCP and GOLEM-3DMCP integrations. It details tool prefixes, API structures, geometry operations, Grasshopper automation, and troubleshooting steps.

1. Overview and Pipeline Value

The Rhino MCP Platform and GOLEM-3DMCP Server connect AI agents to McNeel Rhino 7/8. This allows natural language-driven 3D CAD modeling, NURBS reconstruction, structural planning, and parametric scripting in Grasshopper.

Architecture

  • In-Process socket server: A custom startup script (startup.py) running inside Rhino's Python Script Editor creates a JSON-RPC 2.0 socket server on localhost.
  • MCP Server bridge: A Python MCP server proxy connects the AI assistant to the socket server, translating MCP tool calls to Rhino API calls.
  • Grasshopper runtime: Provides direct control over Grasshopper parameters, enabling real-time solver execution and parameter sweeps.

Which server

ServerUse forFallback when
bdb_rhino_mcp (McNeel router, default version 8)Primary: Rhino 7/8 modelling and Grasshopper control through the McNeel bridge.Undocumented.
bdb_rhino_mcp_fallback (GOLEM-3DMCP, 105 tools)Scene audit, geometry create/boolean/loft, Grasshopper open/set/recompute/bake, viewport capture, Python execution.bdb_rhino_mcp is unavailable (pair labelled primary/fallback for redundancy; no other trigger documented).

2. System Instructions

Workflow Priorities

  1. Scene Audit first: Always run scene.get_document_info or scene.list_layers to understand the scale, active units (e.g. millimeters vs. meters), and current hierarchy.
  2. Object Lifecycle: Many geometry operations (like booleans) consume their input objects and return new GUIDs. Update your working memory with the returned GUIDs and discard the consumed ones to prevent OBJECT_NOT_FOUND errors.
  3. Run on UI Thread: Complex viewport updates or UI-dependent Grasshopper runs require queuing on the Rhino UI thread. The MCP server handles this, but be prepared for a 30s timeout under heavy load.
  4. Grasshopper Workflow: Run grasshopper.open_definition before modifying sliders. Always call grasshopper.recompute to update the solver after changing parameter values.

3. Available Tools and API Parameters

GOLEM-3DMCP provides 105 granular tools across multiple categories.

Scene Intelligence (scene.*)

  • scene.get_document_info(): Returns units, tolerance, layer count, and bounding box.
  • scene.list_layers(): Lists all layers, visibility, lock states, and colors.
  • scene.list_objects(object_type: string, layer?: string): Lists objects with optional type filters (point, curve, surface, brep, mesh, subd, annotation).
  • scene.get_object_info(guid: string): Returns precise control points, layer, and metadata.
  • scene.create_layer(name: string, color?: object): Creates layers (nested with :: syntax).

Geometry Creation (create_*)

  • create_box(width: float, depth: float, height: float, corner_x?: float, corner_y?: float, corner_z?: float)
  • create_sphere(radius: float, center_x?: float, center_y?: float, center_z?: float)
  • create_cylinder(radius: float, height: float)
  • create_nurbs_curve(points: list[list[float]], degree?: int)
  • create_polyline(points: list[list[float]], closed?: boolean)
  • create_text(text: string, location: list[float], height: float)

Geometry Operations (operations.* & surfaces.*)

  • operations.boolean_union(guids: list[string]): Unions closed solids.
  • operations.boolean_difference(guid_a: string, guids_b: list[string]): Cuts guids_b from guid_a.
  • operations.offset_curve(curve_id: string, distance: float)
  • operations.fillet_edge(brep_id: string, edge_indices: list[int], radius: float)
  • surfaces.loft(curve_ids: list[string], loft_type?: string)
  • surfaces.sweep1(rail_id: string, shape_ids: list[string])

Grasshopper Automation (grasshopper.*)

  • grasshopper.open_definition(path: string): Loads a .gh or .ghx definition.
  • grasshopper.get_parameters(): Queries all sliders, panels, and outputs.
  • grasshopper.set_parameter(name_or_id: string, value: any): Modifies sliders or inputs.
  • grasshopper.recompute(): Forces a solution recalculation.
  • grasshopper.get_output_value(name_or_id: string): Reads output component values.

Scripting and Viewports (scripting.* & viewport.*)

  • scripting.execute_python(code: string): Runs arbitrary Python script utilizing Rhino and rhinoscriptsyntax.
  • viewport.capture_to_file(path: string, width?: int, height?: int): Captures the current view.
  • viewport.set_camera(location: list[float], target: list[float]): Points the camera.

4. Code Recipes and Prompt Cookbook

Recipe 1: Parametric Truss Setup in Grasshopper

Open a Grasshopper file, adjust structural dimensions, and extract the generated geometry:

// Step 1: Load file
// Tool: grasshopper.open_definition(path="/Users/<username>/project/truss.gh")

// Step 2: Modify span and height
// Tool: grasshopper.set_parameter(name_or_id="SpanSlider", value=150.0)
// Tool: grasshopper.set_parameter(name_or_id="HeightSlider", value=22.5)

// Step 3: Recompute solver
// Tool: grasshopper.recompute()

// Step 4: Bake outputs
// Tool: grasshopper.bake_output(name_or_id="TrussGeometry", layer="StructuralTrusses")

Recipe 2: Custom NURBS Loft Creation

To generate custom ribs and loft a skin between them using Python scripting:

# Call tool: scripting.execute_python(code="...")
import rhinoscriptsyntax as rs
import math

curves = []
for i in range(5):
    points = []
    for j in range(10):
        x = i * 20.0
        y = j * 10.0
        z = math.sin(j * 0.5) * 15.0 + (i * 2.0)
        points.append([x, y, z])
    
    # Create curve
    crv = rs.AddNurbsCurve(points)
    curves.append(crv)

# Loft skin
loft = rs.AddLoftSrf(curves)
rs.ObjectName(loft, "ProceduralNurbsCanopy")
print(f"Lofted canopy created with ID: {loft}")

5. Troubleshooting and Connection Details

Connection Setup

  • Default Port: 9876 (TCP socket communication)
  • Rhino Startup Script: rhino_plugin/startup.py
  • Verify Connection:
    # Check if server is running on the port
    lsof -i :9876
    

Common Errors and Fixes

  1. RhinoConnectionError: Cannot connect to Rhino plugin at 127.0.0.1:9876

    • Cause: The plugin server is not running inside Rhino.
    • Fix: Open Rhino's Python script editor, open rhino_plugin/startup.py, and run it. To automate, add it to Options -> RhinoScript -> Startup scripts.
  2. GH not available

    • Cause: Grasshopper is installed but has not been loaded in the active session.
    • Fix: Run the Grasshopper command in Rhino once, then restart the GOLEM plugin by running:
      import rhino_plugin.startup as golem
      golem.restart_golem()
      
  3. RhinoTimeoutError

    • Cause: An operation took longer than the default 30-second threshold.
    • Fix: Increase the timeout in your .mcp.json config:
      "env": {
        "GOLEM_TIMEOUT": "120"
      }
      

리뷰

아직 리뷰가 없습니다. 사용해 본 소감을 남겨 주세요.

댓글

로그인 없이 쓸 수 있습니다. 사용 후기, 질문, 작동하지 않은 환경 등을 편하게 남겨 주세요.

아직 댓글이 없습니다. 첫 댓글을 남겨 보세요.

보내면 바로 공개됩니다. 이메일 주소나 전화번호 같은 개인 정보는 쓰지 마세요. URL은 2개까지입니다.

0 / 1,000자

이 양식은 reCAPTCHA로 보호되며 Google 개인정보처리방침 및 서비스 약관이 적용됩니다.

같은 저장소의 스킬

Use when manage multiple local CLI agents via tmux sessions (start/stop/monitor/assign) with cron-friendly scheduling.

한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.

hybridlabor-api/aos62026년 10월 16일 업데이트

Use when a hybrid memory system that provides persistent, searchable knowledge management for AI agents (Architecture, Patterns, Decisions).

한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.

hybridlabor-api/aos62026년 10월 16일 업데이트

Reference for how BDB structures autonomous software engineering work — the seven-node dispatcher graph (Architect, TechLead, UI/UX, Engineering, Media/EventTech, Reviewer, Shipping) that /startcycle-graph actually runs. Use when you need the high-level lifecycle framing without inventing your own process.

한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.

hybridlabor-api/aos62026년 10월 16일 업데이트

Tools are how AI agents interact with the world. A well-designed tool is the difference between an agent that works and one that hallucinates, fails silently, or costs 10x more tokens than necessary. This skill covers tool design from schema to error handling.

한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.

hybridlabor-api/aos62026년 10월 16일 업데이트

Harness patterns for coding agents — memory, permissions, context engineering, delegation, skills, hooks, bootstrap.

한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.

hybridlabor-api/aos62026년 10월 16일 업데이트

agenttrail: live map of a multi-agent build in the browser: which plan component is being worked on, by which agent or harness, what is done and what is stuck. Use when a multi-agent pipeline starts (/startcycle, /startcycle-graph, /teamwork-preview) or after a plan-canvas approve, when the user asks to see what the agents are doing ("live map", "build board", "who is running now"), or by yourself whenever a multi-agent run is under way — no user prompt needed.

한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.

hybridlabor-api/aos62026년 10월 16일 업데이트

hybridlabor-api의 스킬 모두 보기

이 스킬의 문제 신고