An automated design agent that generates professional posters and social media assets directly within your workflow via the visual-rag MCP service.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install posterdesign
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 posterdesign 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 Visual-RAG Design?
Visual-RAG Design is a sophisticated image generation skill designed for Openclaw Skills that bridges the gap between text-based requirements and professional visual assets. By leveraging the Model Context Protocol (MCP), this skill allows AI agents to search for high-quality design templates, map user content into specific slots, and render final PNG images for promotional posters, social media covers (like Xiaohongshu), and activity notices.
Unlike generic text-to-image models, Visual-RAG Design focuses on structured layouts and typography, ensuring that text is legible, correctly formatted, and aesthetically placed. It provides a robust quality control loop where the agent can inspect the final render, adjust element positioning via CSS-like properties, and handle visual overflows before delivering the final product to the user.
Visual-RAG Design Use Cases
- Creating promotional posters for marketing campaigns based on core messaging.
- Generating optimized cover images for social media platforms like Xiaohongshu or Instagram.
- Producing activity notification graphics with specific dates, brand names, and titles.
- Rapid prototyping of design layouts for content creators and brand managers.
- Automated batch generation of branded assets with consistent visual styles.
How Visual-RAG Design Works
- The process begins by calling ensure_services to verify the availability of the MCP design server.
- The agent performs a semantic search using search_templates, utilizing specific query templates (format/audience/message/style) to find the most relevant design layout.
- Upon selecting a template, get_template_spec is called to retrieve technical slot details, including character limits and line-break rules.
- The agent populates the template slots with user-provided information, adhering strictly to max_chars and list_line_count constraints.
- The generate_poster function renders the image, returning both a finished 'render' image and an 'inspect' map for debugging.
- A final acceptance check is performed via a quality report, where the agent handles any text overflows or layout issues before providing the download link.
Visual-RAG Design Setup
To enable the Visual-RAG Design capability within your environment, ensure your agent has access to the visual-rag MCP server.
# Example MCP Server Configuration
# Endpoint: https://syncopated-retractively-anitra.ngrok-free.dev/mcp
# Note: If the domain expires, contact technical support to update the service address.
Ensure that the agent's environment has the necessary permissions to access the /files/ directory for saving and retrieving rendered assets.
Visual-RAG Design Data Schema & Taxonomy
The skill organizes its data using a structured template-and-slot system to ensure visual consistency.
| Field | Description | Constraints |
|---|---|---|
| template_id | Unique integer ID for the design layout | Must be a pure integer (e.g., 689) |
| slots | A dictionary mapping UUIDs to text content | Must match the template spec limits |
| adjustments | Optional CSS-like tweaks (transform, fontSize, top) | Used for fine-tuning layout positioning |
| render_path | The local file path for the final PNG | Format: /files/t[ID]_[UUID]_render.png |
Each template includes a 'Quality Report' indicating status via color codes: Red (Critical Overflow), Yellow (Likely Issue), and Green (Normal).
Visual-RAG Design Advanced Features
- Semantic Template Retrieval: Advanced matching based on target audience and emotional tone rather than just keywords.
- Granular Layout Control: Supports CSS properties like translateX, translateY, and opacity for pixel-perfect adjustments.
- Visual Slot Inspection: Generates color-coded 'inspect' images to identify exactly which element is causing layout breaks.
- Slot Source Validation: Differentiates between 'User-Required' data (brands/dates) and 'AI-Creatable' data (slogans/taglines).
- Multi-Agent Support: Compatible with both vision-capable models (for visual review) and text-only models (via strict character limit adherence).
SKILL.md
Loading
Related Openclaw Skills
Clean Text Formatter
A specialized utility that strips Markdown syntax and normalizes whitespace to transform raw AI-generated content into clean, professional plain text.
SPSS Data Cleaning Assistant
An AI-driven assistant designed to automate the cleaning, validation, and transformation of SPSS datasets for research and analysis.
Texas Electricity Savings Monitor
A high-precision AI agent skill designed to automate Texas residential electricity shopping, plan comparison, and savings monitoring via a strict state machine architecture.
Mapbox MCP DevKit Patterns
A comprehensive integration pattern for connecting Mapbox developer APIs with AI coding assistants to manage styles, tokens, and data validation.
Xiaoka Food Log
An intelligent nutritional logging skill that connects AI agents to the Xiaoka Health API for precise calorie counting and dietary advice.
Lottery Predictor V3.8
A sophisticated machine learning toolkit for predicting lottery outcomes through ensemble modeling and 15-dimensional feature engineering.
Featured*
