メニュー
Authoring skill — create and edit cameras and lights, frame the Scene Viewer, and report active camera/view state through typed tools. Pair with houdini-render for viewport capture and ROP render execution.
Codex または Claude でインストール この Prompt をコピーして Codex、Claude、または他のアシスタントに貼り付けると、Skill ページを確認してインストールできます。
SOC 職業分類に基づく
Pipeline skill — viewport capture/flipbook plus ROP render settings and render execution through typed, runtime-aware tools. Exports report written_files, elapsed time, and separate error/warning diagnostics without hanging the host. Pair with houdini-camera-light to set up cameras and lights first.
Scene skill — read-only inspection of the active Houdini hip file and node graphs. Use when you need frame range, hip path, object-level nodes, or node details. Not for creating geometry, sims, or exports — use authoring or interchange skills.
Authoring skill — inspect and edit Houdini node parameters, parameter templates, spare parameters, and channel expressions through typed atomic tools. Use instead of arbitrary Python for reading/setting parms, adding spare parms, and managing expressions. For node connections use houdini-node-graph.
Pipeline skill — generic, adapter-owned project and shot/package automation: set/query the project root, tag and read portable asset metadata, validate a scene for missing files / bad output dirs / dirty state / cook errors, collect file dependencies into a manifest (no copying), and export a shot/package manifest (frame range, cameras, output nodes, caches, written files). No dependency on private production services.
Scene-management skill — typed Houdini scene lifecycle (new/open/save with dirty-scene safeguards), selection (get/set/find), camera discovery, and bounding-box queries. Use these atomic tools instead of arbitrary Python for everyday scene navigation and file operations. Not for node authoring (use houdini-nodes) or geometry/transform edits (use houdini-object-ops).
Set up dcc-mcp-houdini for an agent or operator: install Houdini Python dependencies with hython, generate MCP host configuration, guide the user through the Houdini package / 123.py autostart load step, and run a first live-tool smoke prompt.
| name | houdini-camera-light |
| description | Authoring skill — create and edit cameras and lights, frame the Scene Viewer, and report active camera/view state through typed tools. Pair with houdini-render for viewport capture and ROP render execution. |
| license | MIT |
| compatibility | dcc-mcp-houdini 0.1+, Houdini 20.5+, dcc-mcp-core 0.19.45+ |
| allowed-tools | Bash Read Write Edit |
| metadata | {"dcc-mcp":{"dcc":"houdini","layer":"domain","stage":"authoring","version":"1.0.0","tags":["houdini","camera","light","hlight","viewport","view","lookthrough","authoring"],"search-hint":"create camera light hlight intensity color exposure focal resolution frame view look through","tools":"tools.yaml"}} |
Typed camera and light authoring for agents. All tools are affinity: main.
View tools (frame_view, get_view_state) are UI-aware: in a headless
hython session they return a structured warnings payload with
framed: false / ui_available: false instead of failing.
When both a node and camera are supplied, frame_view frames the node first
and activates the camera last so the returned active_camera is authoritative.
camera: list_cameras, create_camera, update_camera.view: frame_view, get_view_state.light: create_light, update_light.create_camera(parent_path="/obj", name="shotcam", focal=50, resolution=[1920,1080])create_light(light_type="distant", intensity=2.0, color=[1,0.95,0.9])frame_view(camera_path="/obj/shotcam", node_path="/obj/geo1")get_view_state() → confirm active_camerahoudini-render (capture_viewport / render)Light/camera parameter writes are defensive (set only when the parm exists), so
they tolerate hlight::2.0 vs older hlight and Karma/Mantra camera variants.