# Bio Rfdiffusion Inference

> RFdiffusion 蛋白骨架生成推理 skill。用于执行 OneScience RFdiffusion 的 unconditional backbone generation、motif scaffolding、binder design、symmetric oligomer、partial diffusion，收紧 Hydra contig、input_pdb、checkpoint、num_designs、output_prefix、TRB/trajectory 和后接 ProteinMPNN 校验。

- **Type:** Skill
- **Install:** `agentstack add skill-onescience-ai-oneskills-rfdiffusion-inference`
- **Verified:** Yes — security-reviewed for prompt injection and unsafe behavior
- **Seller:** [onescience-ai](https://agentstack.voostack.com/s/onescience-ai)
- **Installs:** 0
- **Category:** [Agent Skills](https://agentstack.voostack.com/c/agent-skills)
- **Latest version:** 0.1.0
- **License:** MIT
- **Upstream author:** [onescience-ai](https://github.com/onescience-ai)
- **Source:** https://github.com/onescience-ai/OneSkills/tree/master/skills/onescience-role/assets/bio_domain/bio-inference/rfdiffusion-inference

## Install

```sh
agentstack add skill-onescience-ai-oneskills-rfdiffusion-inference
```

Requires the [AgentStack CLI](https://agentstack.voostack.com/docs/cli). Works with Claude Code, Cursor, and any MCP-compatible agent.

## About

# RFdiffusion 推理

## 使用边界

用于生成蛋白骨架或基于目标/基序的骨架设计。若目标是从骨架生成氨基酸序列，读取 `../proteinmpnn-inference/SKILL.md`。

## 可复用资源

- `{onescience_path}/onescience/examples/biosciences/_manifests/model_requests/rfdiffusion_request.yaml`：contig、input PDB、checkpoint、num designs、partial diffusion 和输出模板。
- `references/rfdiffusion_execution.md`：常见模式、参数风险和输出解释。

## 推荐流程

1. 读模型卡：`onescience-coder/assets/models/rfdiffusion.md`。
2. 固定入口：`{onescience_path}/onescience/examples/biosciences/RFdiffusion/scripts/run_inference.py`。
3. 明确模式：unconditional、motif scaffolding、binder、partial diffusion、symmetry。
4. Preflight：contig 长度、input_pdb、hotspot、checkpoint、IGSO3 cache、output_prefix、num_designs。
5. Postflight：PDB、TRB、traj、contig 映射、是否只有 Gly backbone、是否需要 ProteinMPNN。

## 禁止事项

- 不要把 RFdiffusion 输出当成完整侧链设计。
- partial diffusion 必须保证输入结构长度与 contig/被扩散区域一致。
- 不要随意改 `model` / `preprocess` / `diffuser` 强绑定训练参数。

## Source & license

This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.

- **Author:** [onescience-ai](https://github.com/onescience-ai)
- **Source:** [onescience-ai/OneSkills](https://github.com/onescience-ai/OneSkills)
- **License:** MIT

Install and usage instructions live in the source repository linked above.

## Pricing

- **Free** — Free

## Security capabilities

Automated source analysis of v0.1.0 — what this tool can access:

- **Network access:** no
- **Filesystem access:** no
- **Shell / process execution:** no
- **Environment & secrets:** no
- **Dynamic code execution:** no

*"Yes" means the capability is present in the source — more access means more to trust, not that it is unsafe.*


## Versions

- **0.1.0** — security scan: passed — Imported from the upstream source.

## Links

- Listing page: https://agentstack.voostack.com/l/skill-onescience-ai-oneskills-rfdiffusion-inference
- Seller: https://agentstack.voostack.com/s/onescience-ai
- Browse the marketplace: https://agentstack.voostack.com/browse

---
Listed on AgentStack — the marketplace for AI agent skills and MCP servers. Every listing is security-reviewed. Creators keep 70%.