Skip to main content

Fastfold CLI — An autonomous agent for drug discovery research

Project description

Fastfold Agent CLI

Where scientists and AI agents work together doing real science.

Ask questions in natural language. Fastfold CLI plans the analysis, selects the right tools, executes them, validates results, and returns data-backed conclusions. Integrates with Fastfold AI Cloud for GPU compute, protein folding, workflow orchestration, and team collaboration.

Built on top of Claude Agent SDK and CellType CLI.

Prerequisites

  • Python 3.10+ (recommended: let uv install managed interpreters).
  • uv — install from the Astral docs: Installing uv. Quick options:
    • macOS / Linux (standalone installer):

      curl -LsSf https://astral.sh/uv/install.sh | sh
      
    • Windows (PowerShell):

      powershell -ExecutionPolicy ByPass -c "irm https://astral.sh/uv/install.ps1 | iex"
      

      Alternatively: winget install --id=astral-sh.uv -e (see Astral docs for other methods).

      After installing uv, close and reopen your terminal or PowerShell so PATH picks up the uv executable.

Quick install

Requires Python 3.10+ and prerequisites above.

uv tool install "fastfold-agent-cli[all]" --python 3.10

Windows users — prefer WSL2 + Ubuntu ([all]): tiledbsoma does not publish usable native Windows wheels, so [all] on cmd/PowerShell usually fails. Install Windows Subsystem for Linux following Microsoft’s guide: Install WSL (recommended default distro Ubuntu), open an Ubuntu terminal, install uv + Python there, then run uv tool install "fastfold-agent-cli[all]" --python 3.10 inside WSL.

Staying on native Windows cmd/PowerShell: use [win_build] instead of [all] (same stack minus scanpy / cellxgene-census / tiledbsoma), or explicitly. Claude Code ships inside claude_agent_sdk, so uv tool install is normally enough — no npm on Linux/mac. On native Windows, CLI ≥ 0.0.41 validates real launcher spawnability and auto-fixes short launcher cache paths (fastfold setup, fastfold autofix, or interactive /autofix) to avoid WinError 206 from deep site-packages; 0.0.37 also prefers claude on PATH if you install @anthropic-ai/claude-code. Fallbacks: set UV_TOOL_DIR shorter before reinstall, or FASTFOLD_CLAUDE_CODE_CLI.

uv tool install "fastfold-agent-cli[chemistry,biology,ml,analysis]" --python 3.10

Convenience extra (CLI ≥ 0.0.38 on PyPI; Windows uv‑only installs use a short‑path bundled copy automatically):

uv tool install "fastfold-agent-cli[win_build]" --python 3.10

Authentication

# Interactive setup wizard (recommended — configures Anthropic + Fastfold AI Cloud keys)
fastfold setup

# Or set directly
export ANTHROPIC_API_KEY="sk-ant-..."
export FASTFOLD_API_KEY="sk-..."

# Non-interactive (CI/scripting)
fastfold setup --api-key sk-ant-... --fastfold-api-key sk-...

Getting Started

# Start interactive session
fastfold

# Single query
fastfold "What are the top degradation targets for this compound?"

# Validate setup
fastfold doctor

# List available tools
fastfold tool list

# List loaded skills
fastfold skill list

Interactive commands

Inside fastfold interactive mode:

  • /help — command reference + examples
  • /tools — list all tools with status
  • /agents N <query> — run N parallel research agents
  • /sessions, /resume — session lifecycle
  • /copy, /export — save/share outputs
  • /usage — token and cost tracking

Quick examples

Target prioritization

fastfold "I have a CRBN molecular glue. Proteomics shows it degrades
          IKZF1, GSPT1, and CK1α. Which target should I prioritize?"

Protein folding

fastfold "Fold this sequence with boltz-2 and find the binding pockets: MALWMRLLPLL..."

Combination strategy

fastfold "My lead compound is immune-cold. What combination strategy should I use?"

Key Features

190+ Domain Tools

Category Examples
Target Neosubstrate scoring, degron prediction, co-essentiality networks
Chemistry SAR analysis, fingerprint similarity, scaffold clustering
Expression L1000 signatures, pathway enrichment, TF activity, immune scoring
Viability Dose-response modeling, PRISM screening, therapeutic windows
Biomarker Mutation sensitivity, resistance profiling, dependency validation
Clinical Indication mapping, population sizing, TCGA stratification
Safety Anti-target flagging, multi-modal profiling, SALL4 risk
Structure AlphaFold fetch, docking, binding sites, MD simulation
Folding Fastfold AI Cloud: boltz-2, monomer, multimer, simplefold_*
Literature PubMed, OpenAlex, ChEMBL search
DNA ORF finding, codon optimization, primer design, Gibson/Golden Gate assembly

Agent Skills

Fastfold CLI ships with a bundled skill catalog and supports user-installed skills:

fastfold skill list          # see loaded skills

# Install additional skills (requires npx), check https://skills.sh
npx skills addd <owner/repo>

Data Management

fastfold data pull depmap    # DepMap CRISPR, mutations, expression
fastfold data pull prism     # PRISM cell viability
fastfold data pull msigdb    # Gene sets
fastfold data pull alphafold     # Protein structures (on-demand)

# Or point to existing data
fastfold config set data.depmap /path/to/depmap/

Reports

fastfold report list         # list reports
fastfold report publish      # convert latest .md to .html
fastfold report show         # open in browser

Troubleshooting

Symptom Fix
fastfold fails at startup fastfold doctor
No API key fastfold setup or export ANTHROPIC_API_KEY=...
Data not found fastfold data pull <dataset>
tiledbsoma / WinError installing [all] on Windows Prefer WSL2 + Ubuntu and [all] in Linux. Native Windows: [win_build] or [chemistry,biology,ml,analysis] — see Quick install.
Missing dependency (pip fallback) pip install "fastfold-agent-cli[all]"
ModuleNotFoundError: No module named 'termios' (interactive fastfold on Windows) Upgrade fastfold-agent-cli ≥ 0.0.36 (e.g. uv tool install "fastfold-agent-cli[win_build]" --python 3.10 --upgrade).
CLINotFoundError / WinError 206 when spawning Claude (Windows) ≥ 0.0.42: run fastfold autofix (or interactive /autofix) and re-run fastfold doctor. This release also avoids oversized Windows CLI command lines by moving large system instructions into the stdin payload path. If still blocked, set shorter UV_TOOL_DIR (uv tools directory) + reinstall, or set FASTFOLD_CLAUDE_CODE_CLI to a known good global Claude Code launcher path.
Session lost fastfold --continue

Contributing

git clone https://github.com/fastfold-ai/fastfold-agent-cli.git
cd fastfold-agent-cli
uv venv --python 3.12 && uv sync
fastfold setup
pytest tests/

License

MIT — see LICENSE

Credits

Based on CellType CLI

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

fastfold_agent_cli-0.0.42.tar.gz (2.7 MB view details)

Uploaded Source

Built Distribution

If you're not sure about the file name format, learn more about wheel file names.

fastfold_agent_cli-0.0.42-py3-none-any.whl (2.3 MB view details)

Uploaded Python 3

File details

Details for the file fastfold_agent_cli-0.0.42.tar.gz.

File metadata

  • Download URL: fastfold_agent_cli-0.0.42.tar.gz
  • Upload date:
  • Size: 2.7 MB
  • Tags: Source
  • Uploaded using Trusted Publishing? Yes
  • Uploaded via: twine/6.1.0 CPython/3.13.12

File hashes

Hashes for fastfold_agent_cli-0.0.42.tar.gz
Algorithm Hash digest
SHA256 f6aafae612dafe50090a4f70f0e99f3084fd3dcf3206a73ff68cc59f8bba3621
MD5 c6589ba1641ccd7bd7096272e4aee293
BLAKE2b-256 51d9a7063d5aa04f4b140d20c111175cdf51f39b114001706c30059cb2c4c7b8

See more details on using hashes here.

Provenance

The following attestation bundles were made for fastfold_agent_cli-0.0.42.tar.gz:

Publisher: publish.yml on fastfold-ai/fastfold-agent-cli

Attestations: Values shown here reflect the state when the release was signed and may no longer be current.

File details

Details for the file fastfold_agent_cli-0.0.42-py3-none-any.whl.

File metadata

File hashes

Hashes for fastfold_agent_cli-0.0.42-py3-none-any.whl
Algorithm Hash digest
SHA256 58e751e13b90103a9c815e0386fd6e7021303440cf6f048121a212b9a8e1c87e
MD5 0df6a7f8cadb618c8b63e1da065afd1e
BLAKE2b-256 e57073b790997fdaec60761be43585ba6ac3c7246be78cee3bcd136cd8cfea32

See more details on using hashes here.

Provenance

The following attestation bundles were made for fastfold_agent_cli-0.0.42-py3-none-any.whl:

Publisher: publish.yml on fastfold-ai/fastfold-agent-cli

Attestations: Values shown here reflect the state when the release was signed and may no longer be current.

Supported by

AWS Cloud computing and Security Sponsor Datadog Monitoring Depot Continuous Integration Fastly CDN Google Download Analytics Pingdom Monitoring Sentry Error logging StatusPage Status page