A comprehensive automation skill for submitting protein sequences to the FastFold Jobs API and managing the structural biology workflow.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install fastfold-ai-fold
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 fastfold-ai-fold 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 Fold: FastFold Protein Folding?
The Fold skill is a specialized tool designed to interact with the FastFold Jobs API, allowing researchers and developers to automate protein folding tasks. It provides a standardized way to submit sequences, monitor job progress, and retrieve high-fidelity structural data. By integrating this functionality into Openclaw Skills, users can bridge the gap between sequence analysis and 3D structural modeling without manual intervention.
This skill is self-contained, featuring its own OpenAPI schema and pre-configured Python scripts to ensure consistent behavior across different environments. It handles the nuances of the FastFold ecosystem, including support for various models like Boltz-2 and handling different job types from simple monomers to complex multi-chain assemblies.
Fold: FastFold Protein Folding Use Cases
- Submitting protein sequences for high-accuracy folding using the Boltz-2 model.
- Automating the fold-and-wait workflow for large-scale structural biology projects using Openclaw Skills.
- Programmatically retrieving pLDDT and PTM metrics to evaluate the confidence of predicted structures.
- Generating 3D structure viewer links and downloading CIF/PDB artifacts for downstream analysis.
How Fold: FastFold Protein Folding Works
- The skill verifies the presence of a FastFold API key in the local environment or a .env file.
- A folding job is submitted to the /v1/jobs endpoint with specified sequences and parameters using Openclaw Skills.
- The skill initiates a polling sequence, checking the job status at regular intervals until it reaches a terminal state (COMPLETED, FAILED, or STOPPED).
- Once completed, the skill fetches the prediction payload, including the CIF/PDB URLs and associated confidence metrics.
- Users can then download the resulting structural files or access the integrated 3D viewer through generated cloud links.
Fold: FastFold Protein Folding Setup
To get started with this skill within Openclaw Skills, follow these steps:
- Generate an API key from the FastFold dashboard.
- Create a
.envfile at your project root using the provided template:
cp skills/fold/references/.env.example .env
- Add your
FASTFOLD_API_KEYto the.envfile. - Execute folding jobs using the included Python scripts:
python skills/fold/scripts/create_job.py --name "MyProtein" --sequence "MALW..." --model boltz-2
Fold: FastFold Protein Folding Data Schema & Taxonomy
The skill organizes protein data and job metadata using a structured taxonomy:
| Data Type | Description | Format |
|---|---|---|
| Job Input | Sequence data, model parameters, and constraints | JSON |
| Status | Current lifecycle state (PENDING, RUNNING, etc.) | String |
| Artifacts | 3D structural coordinates | CIF / PDB |
| Metrics | Confidence scores (pLDDT, PTM, iPTM) | JSON |
| Viewer Link | Direct URL to the FastFold 3D structure viewer | HTTPS URL |
Fold: FastFold Protein Folding Advanced Features
- Support for complex multi-chain protein folding and ligand sequences.
- Advanced constraint handling for pocket-specific or bond-specific folding requirements.
- High-level automation scripts for batch processing and result polling within Openclaw Skills.
- Flexible model selection including Boltz-2 for state-of-the-art structural predictions.
- Automated .env management and secure credential handling for developer workflows.
SKILL.md
Loading
Related Openclaw Skills
Emergence Codex
A symbolic semantic architecture language designed to enhance AI reasoning and prevent generic LLM output smoothing.
SVG Artist
Generate high-quality vector illustrations and icons directly from text prompts using LLM-driven SVG code generation.
SVG Animator
An innovative animation engine that uses LLMs to generate SVG code sequences and compile them into high-quality MP4 or GIF videos.
Cartoon Pet Generator
A programmatic image generation tool that creates customizable SVG and PNG cartoon pets using Node.js without requiring external AI APIs.
WindSensei
WindSensei is an intelligent wind forecast and spot management assistant designed for kitesurfing, wingfoiling, and wind sports enthusiasts using Openclaw Skills.
Masonry
Masonry is a professional CLI-based tool for generating high-fidelity AI images and videos using advanced text-to-media models.
Featured*
