An automated agent skill that parses text prompts from research papers to generate and save professional scientific diagrams using Gemini and MCP.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install draw-paper
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 draw-paper 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 Draw: Scientific Diagram Generator?
The Draw skill is a specialized automation tool designed for researchers and developers who need to transform text-based paper analyses into high-quality visual assets. By leveraging Openclaw Skills, this agent reads specific input files, extracts diagram prompts, and interacts with the Gemini interface via MCP to produce academic-grade illustrations. It operates as a strict state-machine-driven executor, ensuring every step from directory creation to the final image save is handled with technical precision and detailed error reporting.
This skill is particularly valuable for users who manage large volumes of literature and require a consistent, vector-style aesthetic for deep learning architecture diagrams or complex scientific figures. By integrating this into your workflow through Openclaw Skills, you can automate the tedious process of manual prompting and file management, allowing the AI to handle the end-to-end lifecycle of image creation and storage.
Draw: Scientific Diagram Generator Use Cases
- Automating the generation of neural network architecture diagrams from paper summaries.
- Bulk processing scientific illustration prompts extracted from literature reviews.
- Creating consistent, vector-style academic figures for top-tier conference submissions.
- Managing a local library of AI-generated images indexed by research paper filenames.
How Draw: Scientific Diagram Generator Works
- The agent reads a source text file from the input directory and parses its content to extract prompts identified by Figure or Table keywords.
- A dedicated local output directory is created based on the input filename to organize the generated assets.
- The agent initializes the Gemini interface through MCP, configuring the AI with a professional scientific illustrator persona.
- A controlled loop begins where the agent sequentially sends each extracted prompt to the Gemini model.
- The agent monitors the browser state, waiting for the page to finish loading and for the image to be fully rendered.
- Once ready, the agent triggers the download and saves the image to the local path using a standardized naming convention.
Draw: Scientific Diagram Generator Setup
To use this skill, ensure your environment is configured with the necessary local paths and MCP access.
# Ensure your local directories are prepared
mkdir -p /home/xie/桌面/analysis
mkdir -p /home/xie/桌面/images
# Place your input file (e.g., analysis.txt) in the input directory
# Invoke the skill via your agent by passing the filename parameter
The agent will handle the browser interaction and state management automatically through the Openclaw Skills framework.
Draw: Scientific Diagram Generator Data Schema & Taxonomy
The skill organizes data using a structured filesystem approach to ensure traceability:
| Component | Description | Path/Format |
|---|---|---|
| Input File | Plain text file with prompts | /home/xie/桌面/analysis/{filename} |
| Output Folder | Sub-directory per project | /home/xie/桌面/images/{filename_root}/ |
| Image Files | Generated PNG illustrations | figure_{index}.png |
| State Logs | Real-time execution status | STATE: IMAGE_SAVED_{index} |
Draw: Scientific Diagram Generator Advanced Features
- Strict state machine execution ensures that no generation step is skipped and every action is validated before proceeding.
- Automatic persona initialization ensures Gemini always adopts a professional scientific illustrator role for consistent visual output via Openclaw Skills.
- Robust error handling halts execution immediately upon failure, preventing corrupted data or incomplete image sets.
- Dynamic directory management automatically organizes output based on the input source file, maintaining a clean workspace for large-scale research projects.
SKILL.md
Loading
Related Openclaw Skills
Medical Research Assistant for Paper Analysis
An automated workflow for analyzing medical segmentation papers and generating academic visualization prompts via Gemini Pro.
Screen Video Maker
An AI-powered video editing skill that transforms raw screen recordings into polished, shareable tutorial videos through simple natural language commands.
Cron Job Token Auditor
An expert diagnostic tool designed to audit OpenClaw Gateway cron jobs and identify token-saving opportunities by migrating deterministic tasks to system-level timers.
diff4
A secure CLI tool for encrypting and sharing git diffs and source files using client-side AES-256-GCM encryption.
PRD Skill
An AI-powered assistant designed for Product Managers to generate, optimize, and expand structured Product Requirement Documents (PRDs) from natural language.
GitHub Hosts Optimizer
A networking optimization tool that maps GitHub domains to the fastest available IP addresses in the Windows hosts file for improved connection speeds.
Featured*
