Why use Fastfold

• Scientists + agents: collaborate with AI agents to plan, run, and iterate on folding experiments.
• Managed infrastructure: run jobs on cloud GPUs and containers without handling DevOps yourself.
• Reproducibility first: keep runs versioned and traceable so results are easy to compare and share.
• Production-ready integrations: use REST endpoints directly or ship quickly with the Python SDK and CLI.

Quick start

Here is a quick fold using the Python SDK.

pip install fastfold-ai

from fastfold import Client

client = Client()

my_job = client.fold.create(
    sequence="LLGDFFRKSKEKIGKEFKRIVQRIKDFLRNLVPRTES",
    model="boltz-2",
    is_public=True,
)

results = client.jobs.wait_for_completion(my_job.id, poll_interval=5.0, timeout=900.0)
print("Status:", results.job.status)
print("CIF URL:", results.cif_url())
print("Viewer:", results.get_viewer_link())

This example creates a fold job, waits for completion, then prints the final job status, a direct CIF artifact URL, and a viewer link you can share publicly if is_public=True is set.

SDK and API examples

• Start with the SDK guide: /python-sdk
• Explore practical recipes: /python-examples
• Run from terminal/CI: /cli
• Browse endpoint docs: /api

Skills for agents

If you use AI agents, install the official Fastfold skills package to help agents run fold workflows end-to-end (create job, wait for completion, fetch CIF/PDB, metrics, and viewer links).

npx skills add fastfold-ai/skills

The repository currently includes a fold skill focused on Fastfold Jobs API automation, and supports loading FASTFOLD_API_KEY from .env for local scripting.

• Skills repo: github.com/fastfold-ai/skills
• Skills listing: skills.sh/fastfold-ai/skills
• Detailed guide: /skills

llms.txt

This docs site includes AI-friendly exports:

• /llms.txt: concise index with documentation links.
• /llms-full.txt: full combined markdown content.

You can also append .mdx to docs paths to retrieve markdown content for a specific page (for example, /python-sdk.mdx).

Next steps

Create an API key in the dashboard, run a first fold, then choose your integration path:

• API-first: /api
• SDK/CLI-first: /quick-start