AIGC-Director is an end-to-end AI video production agent that automates the transition from script to screen through a structured, human-in-the-loop workflow.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install aigc-director
2. Manual Installation
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
3. Prompt Installation
Copy this prompt to OpenClaw to install it automatically.
Help me install aigc-director using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Prefer to download?
Get the raw skill files in a ZIP archive.
What is AIGC-Director Agent Skill?
AIGC-Director functions as a persistent intelligent video production assistant, guiding developers and creators through a sophisticated six-phase pipeline. It covers everything from initial script brainstorming to character design, storyboarding, and final video assembly. Built for the Openclaw Skills ecosystem, it runs entirely on local infrastructure using a FastAPI backend and a Next.js frontend to ensure maximum privacy and control.
The core philosophy of this skill is the human-in-the-loop interaction model. Unlike black-box generators, AIGC-Director pauses at nine critical checkpoints to present artifacts and wait for user confirmation. This ensures that every creative decision, from the choice of a plot line to the specific visual style of a character, is validated before moving to the next computationally intensive stage.
AIGC-Director Agent Skill Use Cases
- Automating the full lifecycle of AI video production from a simple text prompt.
- Designing consistent characters and environments for long-form narrative content.
- Creating professional-grade storyboards and reference images for visual storytelling.
- Executing precise image-to-video transformations with granular control over prompts.
- Managing local AI workflows using diverse models like DeepSeek, OpenAI, and DashScope via personal API keys.
How AIGC-Director Agent Skill Works
- Service Verification: The agent checks if the local backend and frontend services are active and verifies the presence of required API keys in the environment.
- Script Generation: The agent brainstorms loglines, expands the creative concept, and structures the final script into acts or scenes.
- Character & Scene Design: It generates visual identities for all key characters and environments involved in the script.
- Storyboarding: Detailed storyboards are created to map out the visual flow of the entire video project.
- Reference Generation: High-quality reference images are produced to serve as the visual foundation for the video generation models.
- Video Synthesis & Assembly: The agent triggers video generation APIs and performs post-editing to compile the final media assets.
AIGC-Director Agent Skill Setup
To deploy this skill within your Openclaw Skills environment, ensure your local servers are initialized:
# Check if backend (port 8000) and frontend (port 3000) are running
curl http://localhost:8000/api/health
# Verify your API configuration
cat aigc-claw/backend/.env | grep -E "API_KEY|KEY"
For first-time setup, execute the full initialization routine found in references/init_project/init_all.md to install dependencies and configure the FastAPI and Next.js environments.
AIGC-Director Agent Skill Data Schema & Taxonomy
AIGC-Director maintains a rigorous organizational structure for all generated assets within the aigc-claw/backend/code/result/ directory:
| Directory | Data Type | Contents |
|---|---|---|
script/ |
Markdown/JSON | Plot lines, expanded scripts, and scene-by-scene breakdowns. |
image/ |
PNG/JPG | Character designs, environmental backgrounds, and visual references. |
video/ |
MP4 | Individual scene clips and the final edited video output. |
Each project is associated with a unique session_id, allowing Openclaw Skills to track the state across multiple phases of the production lifecycle.
AIGC-Director Agent Skill Advanced Features
- Human-in-the-Loop Validation: Integrated stopping points that prevent the agent from proceeding without explicit user approval of creative artifacts.
- Multi-Platform Messaging: Built-in support for pushing project updates and media files directly to Feishu or WeChat.
- Sandbox Workspace: Access to individual LLM, VLM, and text-to-image tools for ad-hoc creative tasks outside the main pipeline.
- Adaptive Model Routing: Support for switching between various providers like AliCloud DashScope, ByteDance Ark, and Kuaishou Kling based on the specific task requirements.
- Local IPv4 Previewing: Automatically generates mobile-friendly preview links using local IP addresses instead of localhost for easier cross-device review.
SKILL.md
Loading
Related Openclaw Skills
CTF Web Exploitation
A comprehensive reference for web security exploitation techniques and payloads designed for CTF challenges.
C-Task-Cleanup
An automated maintenance utility that cleans completed Class-C tasks and extracts high-value insights into long-term memory.
Munger Decision Assistant
A structured decision-making assistant that applies Charlie Munger's mental models to analyze complex investment and strategy scenarios.
auto-evolver
An autonomous evolution engine that balances internal self-optimization with external value creation and automated error recovery.
CTF Reverse Engineering
A comprehensive framework for analyzing binaries, game clients, obfuscated code, and custom virtual machines in competitive security environments.
CTF Cryptography
A comprehensive toolkit of cryptanalysis techniques and one-liners designed for solving Capture The Flag (CTF) cryptography challenges.
Featured*
