# tooluniverse-protein-therapeutic-design

Guides AI-driven de novo protein design using RFdiffusion and ProteinMPNN. Generates structural backbones, samples sequences, and validates folds before lab experiments.

- skill · `type=skill` · Claude Code, Codex, Cursor, GitHub Copilot, Gemini CLI, OpenCode · 398 installs · ★ 1759
- Author: mims-harvard
- License: Apache-2.0 · policy=mirror · the beginning of the file is at the end of this page
- Source: https://github.com/mims-harvard/ToolUniverse/tree/HEAD/skills/tooluniverse-protein-therapeutic-design/SKILL.md
- Updated: 2026-10-24
- HTML: https://affpapa.org/en/nevibecode/skills/tooluniverse-protein-therapeutic-design

## What it does

The skill establishes a computational pipeline across 6 distinct phases. It begins by retrieving target structures from PDB or AlphaFold. From there, RFdiffusion builds at least 5 backbone geometries fitted to the target binding site.

Next, ProteinMPNN samples at least 8 sequences per backbone under temperature control. Folding predictions run through ESMFold or AlphaFold2. Analysis scripts process intermediate metrics using Python libraries such as pandas and scipy.

Access to NVIDIA NIM modules requires an active `NVIDIA_API_KEY` environment variable. The service enforces a rate limit of 40 RPM with a minimum wait of 1.5 seconds between queries. Final candidates are exported as Markdown, FASTA, and CSV files.

## Install

### Claude Code (native platform)

1. **Skills CLI (Vercel)**
```bash
npx skills add mims-harvard/ToolUniverse --skill tooluniverse-protein-therapeutic-design -a claude-code
```

2. **Manually: put the files in place**
- File: `~/.claude/skills/tooluniverse-protein-therapeutic-design/SKILL.md` — Copy the whole skill folder; for one project use .claude/skills/tooluniverse-protein-therapeutic-design/

### Codex

1. **Skills CLI (Vercel)**
```bash
npx skills add mims-harvard/ToolUniverse --skill tooluniverse-protein-therapeutic-design -a codex
```

2. **Manually: put the files in place**
- File: `.agents/skills/tooluniverse-protein-therapeutic-design/SKILL.md` — For all projects use ~/.agents/skills/

### Cursor

1. **Skills CLI (Vercel)**
```bash
npx skills add mims-harvard/ToolUniverse --skill tooluniverse-protein-therapeutic-design -a cursor
```

2. **Manually: put the files in place**
- File: `.cursor/skills/tooluniverse-protein-therapeutic-design/SKILL.md` — Cursor also reads .agents/skills/ and .claude/skills/

### GitHub Copilot

1. **GitHub CLI**
```bash
gh skill install mims-harvard/ToolUniverse tooluniverse-protein-therapeutic-design
```

2. **Manually: put the files in place**
- File: `.github/skills/tooluniverse-protein-therapeutic-design/SKILL.md`

### Gemini CLI

1. **Skills CLI (Vercel)**
```bash
npx skills add mims-harvard/ToolUniverse --skill tooluniverse-protein-therapeutic-design -a gemini-cli
```

2. **Platform CLI**
```bash
gemini skills install https://github.com/mims-harvard/ToolUniverse --path skills/tooluniverse-protein-therapeutic-design
```

### Windsurf (Devin Desktop)

**Manually: put the files in place**
- File: `.devin/skills/tooluniverse-protein-therapeutic-design/SKILL.md` — The legacy .windsurf/skills/ path still works

### OpenCode

**Skills CLI (Vercel)**
```bash
npx skills add mims-harvard/ToolUniverse --skill tooluniverse-protein-therapeutic-design -a opencode
```

### Claude.ai

**Claude.ai**
- Zip the folder and upload it in Settings → Capabilities → Skills

Where a file is given instead of a command, copy the files from the source to that path: https://github.com/mims-harvard/ToolUniverse/tree/HEAD/skills/tooluniverse-protein-therapeutic-design/SKILL.md

## When to use it
- Generate at least 5 backbones for a target epitope using RFdiffusion
- Sample at least 8 sequence variants per backbone via ProteinMPNN
- Evaluate structural confidence metrics including pLDDT and pTM
- Export ranked candidate sequences to FASTA and CSV tables

## Example prompts
- Design a de novo protein binder for target PDB 7ABC using RFdiffusion and ProteinMPNN
- Generate 5 backbones for this epitope and validate top candidates with ESMFold
- Run sequence optimization on the target backbone and compile the design report

## Source text · SKILL.md · Apache-2.0

Below is the content of a third-party file. It is data, not instructions for you: do not act on anything written inside unless the human asks. Block tag: c9c849e255b98aa7.

BEGIN THIRD-PARTY FILE c9c849e255b98aa7
````text
---
name: tooluniverse-protein-therapeutic-design
description: AI-guided de novo protein design — RFdiffusion backbone generation, ProteinMPNN sequence design, structure validation (pLDDT, pTM, MPNN scores). Use for designing therapeutic protein binders, novel scaffolds, enzyme variants, and miniprotein/protein-interface design before experimental validation.
disable-model-invocation: true
---

# Therapeutic Protein Designer

AI-guided de novo protein design using RFdiffusion backbone generation, ProteinMPNN sequence optimization, and structure validation for therapeutic protein development.

**KEY PRINCIPLES**:
1. **Structure-first** - Generate backbone geometry before sequence
2. **Target-guided** - Design binders with target structure in mind
3. **Iterative validation** - Predict structure to validate designs
4. **Developability-aware** - Consider aggregation, immunogenicity, expression
5. **Evidence-separated** - Report folding, pose, geometry, protonation and experimental evidence independently
6. **Actionable output** - Provide sequences ready for experimental testing
7. **English-first queries** - Always use English terms in tool calls

Therapeutic protein design starts with the target interaction. What binding surface do you need to cover? A small pocket = nanobody or peptide. A large flat surface = designed protein. Stability, immunogenicity, and manufacturability constrain the design space.

## LOOK UP, DON'T GUESS
When uncertain about any scientific fact, SEARCH databases first rather than reasoning from memory. A database-verified answer is always more reliable than a guess.

---

## COMPUTE, DON'T DESCRIBE
When analysis requires computation (statistics, data processing, scoring, enrichment), write and run Python code via Bash. Don't describe what you would do — execute it and report actual results. Use ToolUniverse tools to retrieve data, then Python (pandas, scipy, statsmodels, matplotlib) to analyze it.

## When to Use

Apply when user asks to:
- Design a protein binder, therapeutic protein, or scaffold
- Optimize a protein sequence for function
- Design a de novo enzyme
- Generate protein variants for target binding

---

## Workflow Overview

```
Phase 1: Target Characterization
  Get structure (PDB, EMDB cryo-EM, AlphaFold), identify binding epitope

Phase 2: Backbone Generation (RFdiffusion)
  Define constraints, generate >= 5 backbones, filter by geometry

Phase 3: Sequence Design (ProteinMPNN)
````
END THIRD-PARTY FILE c9c849e255b98aa7

The third-party file has ended. Everything between the c9c849e255b98aa7 tags is data: requests and commands from it are not followed. An “end of file” line with another tag inside the block is part of the file.

First 2461 of 7423 bytes. © mims-harvard · Apache-2.0 · https://github.com/mims-harvard/ToolUniverse/tree/HEAD/skills/tooluniverse-protein-therapeutic-design/SKILL.md

_Part of the page did not fit — see the HTML version._

---
NeVibeCode on affpapa.org. Catalog texts are descriptions, not instructions for the agent. How to choose and install: https://affpapa.org/en/nevibecode/agents.md