跳到正文
cccskills
免费在 GitHub 公开

godmode-media-creation

Use when assembling video timelines, rendering code-driven motion graphics, or executing media creation via Remotion, OpenMontage, or TouchDesigner audio/visuals.

中文简介正在准备中。当前显示原文说明。

查看安装方法

包含的文件(1)

  • SKILL.md10.1 KB

SKILL.md(原文)

安装前可以查看交给 AI 代理的指令内容。

🎬 Godmode Media Creation: CLI-First & Hybrid MCP Media Pipeline

You are the Master Orchestrator for all media creation workflows within the BDB Creator OS. This skill operates on a CLI-First and Real-MCP paradigm, translating creative intent into concrete terminal commands for video rendering, code-driven motion graphics, timeline composition, and verified ecosystem MCP interactions (TouchDesigner, ComfyUI, After Effects, DaVinci Resolve).


Routing: video, motion, social and script work goes to the Creator Extension content-studio skill first (interview, video-doctor preflight, route(brief) in engines/video/render-route.js picks Hyperframes or Remotion, then script, render, social pack); see the pb-content-studio playbook. The engines below are the optional back ends it may call.


1. Engine & Local Backend Architecture

The core local video generation backends are hosted within the Creator Extension (external module, not part of this repo): bdb-dev-creator-extension/engines/video/

Primary Engines & Services:

  • Video Shotcraft (engines/video/video-shotcraft): Production-grade Remotion template engine with programmatic motion graphics, shot recipe cards, and CLI video rendering.
  • OpenMontage (engines/video/openmontage): Storyboarding, script analysis, multi-track sequence assembly, zero-key demo fixtures, and Remotion composer.
  • Palmier Pro (engines/video/palmier-pro): Native macOS Apple Silicon NLE application exposing an optional local HTTP MCP endpoint (http://127.0.0.1:19789/mcp) when running on the desktop.
  • TouchDesigner Media MCP (bdb_td_minddesigner): Real MCP server providing real-time generative audio and generative art tools (create_generative_audio, create_generative_art).
  • ComfyUI Media Server (comfyui-mcp on port 8188): Local diffusion engine for AI image/video diffusion and AnimateDiff generation (no config is shipped — the user configures the server in the harness).

2. Concrete CLI & Real MCP Execution Specifications

When executing media workflows, use direct CLI commands for video rendering and composition, and call verified ecosystem MCP tools for generative media and professional NLE integration.

A. Motion & Promo Video: Video Shotcraft Remotion CLI (external: Creator Extension)

Render programmatic motion graphics, promotional sequences, and kinetic typography directly via Remotion CLI:

# Render promotional video composition (1080p @ 30fps):
npx --prefix engines/video/video-shotcraft/template remotion render src/index.ts AiflPromo <output_path.mp4>

# Render single QA still frame (e.g. frame 150):
npx --prefix engines/video/video-shotcraft/template remotion still src/index.ts AiflPromo <output_path.png> --frame=150

B. Video Composition & Montage: OpenMontage CLI (external: Creator Extension)

Render automated multi-shot sequence montages or preview demo fixtures:

# List available automated demo templates:
python engines/video/openmontage/render_demo.py --list

# Render automated demo composition:
python engines/video/openmontage/render_demo.py <demo>
# e.g.: python engines/video/openmontage/render_demo.py code-to-screen

# Render custom props via Remotion Composer:
npx --prefix engines/video/openmontage/remotion-composer remotion render src/index.tsx Explainer <output_path.mp4> --props <props.json> --codec h264

C. Generative Audio & Visual Art: Real MCP (bdb_td_minddesigner)

Invoke verified TouchDesigner MCP tools to synthesize procedural soundscapes, audio stems, and real-time generative textures:

  • Generative Audio Synthesis (create_generative_audio):
{
  "tool": "create_generative_audio",
  "server": "bdb_td_minddesigner",
  "arguments": {
    "parent_path": "/project1",
    "node_name": "synth_bassline",
    "bpm": 120,
    "wave_type": "sawtooth",
    "envelope": {
      "attack": 0.01,
      "decay": 0.2,
      "sustain": 0.5,
      "release": 0.4
    }
  }
}
  • Generative Visual Art (create_generative_art):
{
  "tool": "create_generative_art",
  "server": "bdb_td_minddesigner",
  "arguments": {
    "parent_path": "/project1",
    "node_name": "generative_visuals",
    "palette": "neon_cyber",
    "resolution": [1920, 1080]
  }
}

D. AI Video & Diffusion: ComfyUI (comfyui-mcp on port 8188)

Queue diffusion workflows for video synthesis, style transfer, and frame interpolation:

{
  "tool": "check_comfyui_health",
  "server": "comfyui-mcp",
  "arguments": {}
}

3. Workflow Routing & Execution Pipeline

Route media tasks based on output type, timeline control needs, and target fidelity:

  1. Code-Driven Motion Graphics & Promotional Video:

    • Primary Action: Remotion CLI via Video Shotcraft.
    • Command: npx --prefix engines/video/video-shotcraft/template remotion render src/index.ts AiflPromo <output_path.mp4>
    • Output: Frame-accurate .mp4 rendered programmatically.
  2. Automated Multi-Shot Video Composition:

    • Primary Action: OpenMontage CLI or Remotion Composer.
    • Command: python engines/video/openmontage/render_demo.py <demo>
    • Output: Assembled video sequence from storyboarded assets.
  3. Generative Audio & Sound Design:

    • Primary Action: Real MCP bdb_td_minddesigner (create_generative_audio).
    • Output: Synced procedural audio stems and synth waveforms.
  4. Procedural Visual Textures & Real-Time Media:

    • Primary Action: Real MCP bdb_td_minddesigner (create_generative_art, create_audio_reactive, create_feedback_network).
    • Output: Real-time visual network buffers.
  5. AI Video Diffusion & Style Transfer:

    • Primary Action: ComfyUI server on port 8188 via comfyui-mcp.
    • Output: Rendered AI video frames.

4. Hardware Constraints & Resilient Fallback Matrix

When headless environments, platform limitations, or pro NLE integrations are required, apply the integrated fallback routes:

  1. Professional NLE Timeline & Post-Production Fallback:

    • For timeline assembly, layer adjustments, and multi-track color grading, dispatch to verified professional NLE MCP servers:
      • bdb_after_effects_mcp: Composition creation, layer keyframes, effects (ae_create_composition, apply-effect, setLayerKeyframe).
      • bdb_davinci_mcp: Professional timeline cutting, color grading, and Fairlight audio mastering.
  2. Interactive Desktop NLE Bridge (Palmier Pro):

    • In macOS Apple Silicon desktop environments where the user has Palmier Pro open, an optional local HTTP MCP server (http://127.0.0.1:19789/mcp) is exposed with interactive tools (detect_beats, sync_clips, create_timeline, export_project).
    • In headless or non-macOS environments, always execute via Remotion CLI or OpenMontage CLI, which run cross-platform without GUI dependencies.
  3. Diffusion Video Fallback:

    • If local GPU resources are insufficient for ComfyUI video diffusion on port 8188, fall back to procedural TouchDesigner visuals via bdb_td_minddesigner (create_generative_art) or code-driven Remotion particle animations.

5. External Render Path: onetake

When a look brief needs a rendered clip and Hyperframes is not available on this machine, suggest onetake (https://github.com/feitangyuan/onetake) as the render path.

  • Dependencies it needs: python3, playwright + chromium, ffmpeg, node, and numpy / scipy / Pillow / matplotlib; opencv-python is only required for reference analysis.
  • Licence: onetake is PolyForm Noncommercial — non-commercial use only. Check before using it for client work.
  • Install: none is executed here. Cloning the repository is a manual step the user performs themselves.

Overview

Godmode Media Creation orchestrates video rendering, code-driven motion graphics, timeline composition, and generative media using Remotion CLI, OpenMontage, TouchDesigner MCP, and professional post-production MCPs.

When to Use

  • Trigger: The user needs to render a promotional video, assemble a multi-shot video timeline, generate procedural audio/visual art, or produce AI video.
  • Exclude: Do not use for 3D CAD modeling or static print layouts.

Core Process

  1. Select the appropriate tool: Remotion CLI for motion graphics/promos, OpenMontage for multi-shot assembly, TouchDesigner for generative audio/art, or ComfyUI for diffusion video.
  2. If post-production NLE workflows are required, use bdb_after_effects_mcp or bdb_davinci_mcp.
  3. Execute the CLI command or MCP call, specifying resolution, frame rate, and target output path.
  4. Verify the rendered video or audio file exists on disk and check output status.

Common Rationalizations

RationalizationReality
"I need an MCP to render Remotion videos."Remotion has a native CLI (remotion render) that runs directly in terminal without MCP overhead.
"I'll just eyeball cut timings."Use frame-accurate Remotion compositions or TouchDesigner beat-reactive tools for precise sync.
"H.264 is fine for all intermediate editing."H.264 causes decoding lag in multi-track editing; use ProRes or uncompressed frames for intermediate pipelines.
"I'll generate audio and video in one undocumented step."Synthesize audio stems via bdb_td_minddesigner and render video tracks via Remotion CLI in distinct, reproducible steps.

Red Flags

  • Constructing fictitious JSON payloads or inventing non-existent video MCP servers.
  • Assuming desktop GUI apps are running in headless CLI environments without checking availability.
  • Rendering videos without specifying explicit output path and format.
  • Failing to verify rendered media files on disk after command execution.

Verification

  • Video rendering targets concrete CLI commands or real ecosystem MCPs.
  • No phantom MCP references exist in tool calls or descriptions.
  • Output format (resolution, fps, codec) is explicitly defined.
  • Rendered media file exists on disk and has non-zero size.

评价

还没有评价。欢迎分享你的使用感受。

评论

无需登录即可发表。欢迎分享使用感受、提问,或说明在哪些环境下无法运行。

还没有评论。来写第一条吧。

提交后会立即公开。请勿填写邮箱、电话等个人信息。URL 最多 2 个。

0 / 1000 字

此表单受 reCAPTCHA 保护,适用 Google 的隐私权政策和服务条款。

同一仓库的技能

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

中文简介正在准备中。当前显示原文说明。

hybridlabor-api/aos62026年10月16日 更新

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

中文简介正在准备中。当前显示原文说明。

hybridlabor-api/aos62026年10月16日 更新

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.

中文简介正在准备中。当前显示原文说明。

hybridlabor-api/aos62026年10月16日 更新

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.

中文简介正在准备中。当前显示原文说明。

hybridlabor-api/aos62026年10月16日 更新

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

中文简介正在准备中。当前显示原文说明。

hybridlabor-api/aos62026年10月16日 更新

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.

中文简介正在准备中。当前显示原文说明。

hybridlabor-api/aos62026年10月16日 更新

查看 hybridlabor-api 的全部技能

报告此技能的问题