fastfold-ai-fold
by juliocesar-io
Browser & AutomationBrief
Submits and manages FastFold protein folding jobs via the Jobs API.
Description
What This Skill Does
Manages protein structure prediction jobs through the FastFold Jobs API. Handles the full lifecycle: authenticate, submit a fold job, poll until complete, then retrieve CIF/PDB files, quality metrics, and 3D viewer links. Works with single proteins, multi-chain complexes, RNA, DNA, and ligand sequences.
The bundled scripts handle polling, error states, and result parsing so you don't have to write that boilerplate yourself.
When to Use It
- Folding a novel protein sequence to predict its 3D structure
- Automating batch fold submissions and waiting for results in a script
- Downloading CIF files from completed jobs for use in molecular visualization tools
- Checking pTM and ipTM confidence scores after a multi-chain complex prediction
- Getting a shareable FastFold viewer link for a completed fold job
Example Workflow
User asks: Fold this protein sequence MALWMRLLPLLALLALWGPDPAAAFVNQHLCGSHLVEALYLVCGERGFFYTPKT and give me the viewer link
- 1
Check that FASTFOLD_API_KEY is set in the environment or .env file
- 2
Run create_job.py with the sequence and a job name to POST to /v1/jobs
- 3
Run wait_for_completion.py with the returned job ID, polling every 5 seconds until status is COMPLETED
- 4
Run fetch_results.py to retrieve meanPLLDT, ptm_score, and artifact URLs
- 5
Run get_viewer_link.py to build the FastFold cloud viewer URL
Job completed with meanPLLDT 87.3, ptm_score 0.81. CIF available at the signed URL. Viewer: https://cloud.fastfold.ai/mol/new?from=jobs&job_id=<uuid>
Requirements
- FASTFOLD_API_KEY — generate at https://cloud.fastfold.ai/api-keys
Installation
clawhub install juliocesar-io/fastfold-ai-foldRun the CLI command, or paste the prompt into Claude Code, Cursor, Codex or whichever coding agent you use.