Claude Skill MartBrowse skillsWhat is a Skill?
 
← Back to list

Tamarind Bio Molecular Design Cloud

Run open-source structural-biology tools like AlphaFold, RFdiffusion and DiffDock on Tamarind Bio's managed cloud GPUs via REST API or MCP.

Web & APIAdvanced★ 33,030⑂ 3,248AI score 8/10Last updated: Aug 9, 2026

What it does

Tamarind Bio is a cloud platform that runs hundreds of open-source computational-biology tools on managed GPUs: structure prediction (AlphaFold, Boltz-2, Chai-1, ESMFold), protein/binder/de novo design (RFdiffusion, ProteinMPNN, BoltzGen, BindCraft), antibody and nanobody design plus developability, protein–ligand docking (DiffDock, AutoDock Vina), binding-affinity prediction, MSA generation, and molecular dynamics.

This skill teaches Claude to drive that platform correctly:

  • Authenticate with the x-api-key header, reading the key from TAMARIND_API_KEY (never hardcoded)
  • Follow the canonical loop: discover → get schema → validateJob → submit → poll → download results
  • Choose between REST (/tools, /submit-job, /submit-batch, /jobs, /result, /upload) and the hosted MCP server at mcp.tamarind.bio/mcp
  • Avoid the non-obvious foot-guns: /jobs?jobName= returns a bare row (no jobs wrapper), /result is a two-step presigned download, batches must be polled on batchStatus, and a plain string in a file parameter is treated as inline file content
  • Chain jobs (design → fold → score), submit batches of up to 100 jobs, and cap spend with weightedHoursBudget

Who it's for

  • Researchers and grad students who need AlphaFold/Boltz runs without local GPUs
  • Biotech teams designing binders, antibodies or nanobodies with automated verification pipelines
  • Computational-biology engineers screening dozens to hundreds of sequences programmatically
  • Anyone who prefers conversational job submission over hand-writing API calls

Examples

  1. "Fold this sequence with AlphaFold" → Claude checks the tool schema from /tools, submits the job, polls every 30 seconds, downloads the results zip and summarises pLDDT/pTM.
  2. "Design 50 EGFR binders with ProteinMPNN and fold-verify all of them" → runs the design job, then uses MCP submitBatch(fromJob=...) to fold every generated sequence and polls the batch parent's batchStatus for the aggregated output.
  3. "Dock this SDF ligand into the GLP1R_ECD.pdb I uploaded" → uploads via presigned URL, references it by bare filename, submits DiffDock or AutoDock Vina, and pulls getJobLogs if the job stops.

· · · Install guide · · ·

Install in the Claude app (no terminal)
  1. Download the ZIP with the button below.
  2. In Claude, open Settings → Capabilities and turn on 'Code execution and file creation'. (one time)
  3. Go to Customize → Skills → + → 'Upload a skill' and upload the ZIP.
↓ Download ZIP
Install in Claude Code

Let Claude do it — paste this into Claude Code

Install the skill I found on Claude Skill Mart.
Copy the skills/tamarind folder from the GitHub repo K-Dense-AI/scientific-agent-skills into my ~/.claude/skills/tamarind/.
When it's done, tell me in one line what this skill can do.

Install with a command instead

git clone https://github.com/K-Dense-AI/scientific-agent-skills.git && mkdir -p ~/.claude/skills && cp -r scientific-agent-skills/skills/tamarind ~/.claude/skills/

⚠ This is a third-party skill. Check the source repository before installing.

  1. Create a Tamarind Bio account at app.tamarind.bio and generate an API key in the account/API settings (10 free jobs included).
  2. Open a terminal and clone the skills repo: git clone https://github.com/K-Dense-AI/scientific-agent-skills.git
  3. Copy the skill into Claude's skills folder: mkdir -p ~/.claude/skills && cp -r scientific-agent-skills/skills/tamarind ~/.claude/skills/
  4. Export your key: export TAMARIND_API_KEY="your_api_key" (add it to ~/.zshrc or ~/.bashrc to persist).
  5. Make sure you have Python 3.10+ and the deps: pip install requests python-dotenv
  6. Verify connectivity: curl https://app.tamarind.bio/api/tools -H "x-api-key: $TAMARIND_API_KEY" — a JSON list of tools means you're set.
  7. Restart Claude Code and try "Use Tamarind to fold this protein sequence". Optional: if your host supports MCP, register https://mcp.tamarind.bio/mcp with the X-API-Key header for a smoother agentic workflow.
View source on GitHub ↗License: MIT