artclaw_bridge

ArtClaw Tool Creator - AI 引导创建自定义 DCC 工具。支持三种创建方式：包装 Skill、编写脚本、组合工具。

ArtClaw Tool Executor - 执行 ArtClaw Tool Manager 中的工具。当用户请求运行工具时，根据工具类型选择正确的执行方式。

Capture UE viewport screenshots (color, depth, normal buffers) and build AI analysis context. Use when AI needs to: (1) take a screenshot of the current viewport, (2) capture depth/normal maps alongside color, (3) analyze scene composition/lighting/performance visually, (4) get viewport image + camera transform for scene analysis. NOT for: runtime game screenshots, cinematic rendering, Sequencer captures.

ComfyUI 缺失自定义节点自动检测与安装。 Use when AI needs to: (1) install missing custom nodes for a workflow, (2) check if required nodes are available, (3) search and install node packages from ComfyUI registry. ComfyUI only (run_python).

从 SD 内置 30 个 PBR 材质逆向分析的材质配方库。包含 7 个类别配方 + 3 个通用功能配方， 涵盖混凝土/金属/瓷砖/砖墙/布料/木材/有机物的节点图结构、管线设计和制作逻辑。 Use when AI needs to: (1) choose appropriate nodes for a material type, (2) reference professional material graph structure, (3) understand common SD material building patterns, (4) select texture sources for different material categories, (5) plan a material pipeline before building. Substance Designer only (run_python).

Manage ArtClaw skills: list, execute, install, uninstall, update, publish, rename, and create new skills. Use when AI needs to: (1) install/uninstall/update skills, (2) publish user skills to source repository, (3) list available skills and their info, (4) create new skills with AI-generated code, (5) rename or reorganize skills. All operations via Python API in DCC (run_ue_python or run_python). NOT for: OpenClaw CLI skill commands, ClawHub marketplace publishing.

Unity Editor AI agent context and routing skill for ArtClaw Bridge. Use when AI needs to: (1) understand how to operate Unity Editor via ArtClaw MCP, (2) determine which Unity sub-skill to load for a task, (3) get project and environment context (Unity version, project path, connection status), (4) understand the run_unity_python execution model and C# code format. NOT for: direct Unity operations — load the specific skill (unity-scene-ops, unity-gameobject-ops, unity-asset-ops, unity-component-ops, unity-editor-control) for actual operations.

Unity Editor asset management via run_unity_python: import, find, create, move, and delete assets. Use when AI needs to: (1) search assets by type or name (AssetDatabase.FindAssets), (2) import external files, (3) instantiate prefabs in scene, (4) create prefabs from GameObjects, (5) create materials and assign shaders, (6) move/rename/delete assets, (7) refresh AssetDatabase, (8) get asset info (path, GUID, type). NOT for: modifying component properties (use unity-component-ops), scene hierarchy (use unity-gameobject-ops).

Unity Editor component management: add, remove, get, and modify components on GameObjects via run_unity_python. Use when AI needs to: (1) add components (Rigidbody, Collider, Light, Camera, AudioSource, etc.), (2) remove components, (3) get component property values, (4) set component property values (serialized fields), (5) list all components on a GameObject. NOT for: creating GameObjects (use unity-gameobject-ops), script creation (use unity-editor-control).

Unity Editor state control via run_unity_python: undo/redo, play mode, build, console, menu execution, and editor preferences. Use when AI needs to: (1) undo or redo editor operations, (2) enter/exit Play Mode, (3) trigger builds (Build Player), (4) read Unity console logs, (5) execute editor menu items, (6) refresh the project, (7) get editor/project preferences, (8) open editor windows or inspector. NOT for: scene operations (use unity-scene-ops), asset management (use unity-asset-ops).

Unity Editor GameObject creation, deletion, transform, and hierarchy operations via run_unity_python. Use when AI needs to: (1) create/delete/duplicate GameObjects, (2) set position/rotation/scale, (3) get or set parent-child relationships, (4) set active state, (5) get current editor selection, (6) rename GameObjects, (7) move objects in hierarchy. NOT for: adding/removing components (use unity-component-ops), asset operations (use unity-asset-ops).

Unity Editor scene operations via run_unity_python MCP tool. Use when AI needs to: (1) get scene info or hierarchy, (2) open/save/create scenes, (3) find GameObjects by name, tag, or layer, (4) manage Build Settings scene list, (5) query scene statistics (object count, dirty state, path). NOT for: runtime scene loading (Application.LoadScene), asset creation, component editing.

SD 节点创建、连接、参数设置指南。包含原子节点和库节点的正确使用方式。 Use when AI needs to: (1) create nodes, (2) connect nodes, (3) set parameters, (4) build material graphs. Substance Designer only (run_python).

Substance Designer 操作通用规则。所有 SD 图操作前必读。 Use when AI needs to: (1) perform any SD graph modification, (2) check post-operation best practices, (3) understand SD API constraints. Substance Designer only (run_python).

UE5.6/UE5.7 Blueprint graph workflow for feature implementation, input events, node wiring, and graph validation. Use when requests involve adding Blueprint logic, keyboard input behavior, function chains, event graph edits, or pin-level connection guidance.

ArtClaw official Material node graph write operations for UE5. Create, delete, connect, and edit MaterialExpression nodes in a Material's node graph, create new Material assets, set material properties, and recompile materials. Use when AI needs to: (1) build materials from scratch by assembling expression nodes, (2) modify existing material node graphs, (3) connect nodes to material property inputs (BaseColor, Roughness, Normal, etc.), (4) set material-level properties (BlendMode, ShadingModel, TwoSided), (5) help users author procedural or texture-based materials. Works with Material assets (not MaterialInstance — use material_ops.py tools for instance parameter overrides). Requires UE Editor running with ArtClaw plugin. NOT for: reading node graphs (use get-material-nodes skill), material instance parameter queries (use get_material_parameters tool).

通用 DCC 节点图工作流：适用于所有节点图操作（UE 蓝图、SD 材质、UE 材质）。 核心思路：先写伪代码，再映射节点，再创建连接。 Use when AI needs to: build any node graph (Blueprint/SD/UE Material) from scratch or add complex logic to an existing graph. Activate BEFORE the DCC-specific skill to plan the approach, then switch to the specific skill for API details.

ComfyUI CivitAI 集成：搜索、下载 CivitAI 上的模型。 Use when AI needs to: (1) search models on CivitAI, (2) download models from CivitAI, (3) browse popular models. ComfyUI only (run_python).

查询 ComfyUI 上下文：系统信息、模型列表、队列状态、节点类型、工作流历史。 Use when AI needs to: (1) get system info, (2) list available models, (3) check queue status, (4) query node types and schemas, (5) get workflow history. ComfyUI only (run_python).

ComfyUI ControlNet 工作流：Canny、OpenPose、Depth、Lineart 等控制生成。 Use when AI needs to: (1) control image generation with reference image, (2) use ControlNet preprocessors and models, (3) build ControlNet workflows. ComfyUI only (run_python).

ComfyUI 高清修复工作流：低分辨率生成 → 放大 → 重采样。 支持多种放大算法和重采样策略。 Use when AI needs to: (1) upscale generated images, (2) perform high-resolution fix, (3) enhance image details. ComfyUI only (run_python).

ComfyUI 图生图工作流构建与执行。 Use when AI needs to: (1) generate image from another image, (2) perform image-to-image transformation, (3) apply style transfer or enhancement to existing image. ComfyUI only (run_python).

ComfyUI 局部重绘（Inpainting）：修改图像的特定区域。 Use when AI needs to: (1) modify part of an image, (2) remove or add objects, (3) change clothing or background. ComfyUI only (run_python).

ComfyUI 模型管理：查询、下载、删除、搜索模型。 支持 Checkpoints, LoRA, VAE, ControlNet 等所有模型类型。 Use when AI needs to: (1) list available models, (2) download models from CivitAI or HuggingFace, (3) manage model files. ComfyUI only (run_python).

ComfyUI 操作通用规则。所有 ComfyUI 操作前必读。 Use when AI needs to: (1) perform any ComfyUI operation, (2) check post-operation best practices, (3) understand ComfyUI API constraints. ComfyUI only (run_python).

ComfyUI 文生图标准工作流。checkpoint + CLIP + KSampler + VAEDecode + SaveImage。 Use when AI needs to: (1) generate images from text prompts, (2) tune sampling parameters, (3) add LoRA to generation pipeline. ComfyUI only (run_python).

ComfyUI Workflow 动态构建指南。节点类型、连接语法、常用管线模板。 Use when AI needs to: (1) build workflow JSON dynamically, (2) understand node connections, (3) use txt2img/img2img templates. ComfyUI only (run_python).

ComfyUI Workflow 模板库管理：发现、使用、管理、发布 Workflow 模板。 支持官方模板、社区模板、个人模板三级体系。 Use when AI needs to: (1) find and use workflow templates, (2) manage personal workflow library, (3) publish workflows to community. ComfyUI only (run_python).

ComfyUI Workflow 自动诊断与修复。检测缺失节点、参数错误、连接问题， 并提供自动修复方案或手动修复指导。 Use when AI needs to: (1) fix a broken workflow, (2) migrate workflow to new ComfyUI version, (3) optimize workflow structure. ComfyUI only (run_python).

验证 ComfyUI Workflow 的节点是否可用。 在提交 workflow 前检查所有节点类型是否存在。 Use when AI needs to: (1) validate workflow before submission, (2) check if nodes are available, (3) debug workflow issues. ComfyUI only (run_python).

Analyze game scene concept art / screenshots into structured JSON for 3D reconstruction. Extracts objects, spatial layout, camera parameters, lighting, perspective lines, and depth layers. Use when: (1) user provides a scene image and wants structured analysis for 3D scene building, (2) converting 2D concept art to UE/3D engine scene layout, (3) extracting object lists, positions, and spatial relationships from game art. Triggers: "分析场景", "场景原画分析", "scene analysis", "图片分析", "结构化分析", "2D to 3D".

Read a master material's blueprint, extract all parameters (scalar, vector, texture, static switch), and generate a comprehensive Markdown documentation file. Use when AI needs to: (1) document material parameters for artists, (2) generate technical reference for master materials, (3) create handoff documentation for material setups. Saves output to Saved/MaterialDocs/. Only reads material data (no modifications). Requires UE Editor running with ArtClaw plugin.

查询 Blender 编辑器上下文：当前选择、场景信息、文件名、视口相机状态、集合结构。 Use when AI needs to: (1) get selected objects, (2) get scene overview, (3) check current file, (4) inspect viewport state, (5) get collection hierarchy. Blender only (run_python).

Search ArtClaw local knowledge base for UE/Maya/Max API docs, code examples, naming conventions, and best practices. Use when AI needs to: (1) look up correct API usage before writing code, (2) find code examples for specific operations, (3) check naming conventions or project standards. Replaces old MCP tool: knowledge_search.

Manage ArtClaw's memory system: store/get/search/list/delete user preferences, operation history, project conventions, and user habits. Use when AI needs to: (1) remember user preferences or conventions, (2) check operation history (what was done before), (3) store lessons learned from errors, (4) maintain and compact memory tiers. Replaces old MCP tool: memory (7 actions: store/get/search/list/delete/check_operation/maintain).

Query DCC editor context: current selection, scene info, level/file name, viewport camera, object counts. Use when AI needs to: (1) get selected objects, (2) get scene overview, (3) check current level/file, (4) inspect viewport state. Works with UE (run_ue_python), Maya and Max (run_python). Replaces old MCP tools: get_editor_context, get_selected_objects, get_scene_info.

SD FX-Map 节点使用指南。FX-Map 是 SD 中最强大的图案生成节点， 能实现所有基于网格的重复/散布/迭代图案。tile_generator/splatter/cells 等核心库节点 都是 FX-Map 的封装。包含：节点类型、迭代逻辑、函数图参数化、常用图案模式。 Use when AI needs to: (1) create custom tile/scatter patterns, (2) build iterative procedural patterns, (3) understand how tile_generator works internally, (4) implement patterns that library nodes cannot achieve. Substance Designer only (run_python).

SD 纹理/图案生成器库节点深度分析。覆盖 21 个核心生成器的内部引擎、参数、适用场景。 分为三类：图案生成器(tile_generator/splatter等)、噪波生成器(perlin/cells/voronoi等)、 特征纹理生成器(moisture_noise/wood_fibers等)。 Use when AI needs to: (1) choose the right generator for a material, (2) understand generator parameters for fine-tuning, (3) know which generators are FX-Map vs Pixel Processor based, (4) build custom patterns by combining generators. Substance Designer only (run_python).

捕获 SD 节点输出纹理用于 AI 视觉分析。支持按节点 ID、按输出通道批量截图。 Use when AI needs to: (1) see what a node's output looks like, (2) capture all PBR outputs for quality review, (3) inspect intermediate node results to guide optimization, (4) visually verify material correctness after building a graph. NOT for: final production export, 3D viewport screenshot. Substance Designer only (run_python).

SD 节点选型目录 + 动态查询。包含库节点的 sbs 文件名、输出端口、用途分类， 供 AI 选择合适的纹理源。详细参数和端口通过动态查询获取（永远准确）。 Use when AI needs to: (1) find the correct sbs filename for a library node, (2) choose appropriate noise/texture generators for a material, (3) dynamically query node params/ports before setting values. Substance Designer only (run_python).