onescience-mlmd-campaign-orchestrator SKILL.md skillA
onescience-mlmd-campaign-orchestrator from onescience-ai/oneskills is an agent skill: a SKILL.md folder an AI agent loads when a task calls for it.
mdr add onescience-ai/oneskills/onescience-mlmd-campaign-orchestrator@git:20261103.d90dac1Pins git:20261103.d90dac1 in mdr.lock. Pin the fingerprint instead.
- Audit grade
- A, 17 of 17 checks passed
- Kind
- SKILL.md skill
- Repository
- onescience-ai/oneskills
- Path
skills/onescience-mlmd-campaign-orchestrator/SKILL.md- Versions recorded
- 5
- Latest version
- git:20261103.d90dac1, committed Nov 3, 2026
- First version recorded
- Oct 15, 2026
- Size
- 11,231 bytes
- Repository stars
- 24
- License
- MIT
端到端编排“DFT 结构优化 → AIMD 出轨迹 → MACE 数据集 → MACE 势函数训练 → LAMMPS 长时 MD”的机器学习分子动力学流水线。当用户用一句话要求“给 <界面体系> 做一套 MLMD：先 VASP 做 DFT 优化，再跑 VASP AIMD 出轨迹，转成带能量/力的 MACE extxyz 数据集（按时间块切 train/valid/test，不要随机打散），训一个 MACE 势函数（力误差达标），最后导出 LAMMPS 模型跑长时 MD，把皮秒 AIMD 推到纳秒”时加载。负责把自然语言变成结构化 campaign 规格、逐阶段执行并设硬门限、不达标即停并报告。.
Collected from public GitHub. Every version is saved exactly as it was and never changes. Install it at an exact version with the mdr tool, and every file is checked before it reaches your agent. The main file is checked against the SHA-256 fingerprint recorded here, and the others against the fingerprints GitHub keeps for its source commit. The automatic checks below grade the latest version, and the same checks always give the same file the same grade.
What the file says
# Skill: onescience-mlmd-campaign-orchestrator
## 这个技能解决什么
把“给我做一套 MLMD”这类**一句话需求**，变成一条**可复现、带门限、可交接**的流水线，
并把科研要害固化为默认规则：**中性 vs 恒电位**、**时间块切分（防泄漏）**、**误差验收门限**、
以及“**不达标就停、不擅自改科学参数**”。
## When to Use
用户提出类似（不限于此）：
> “我需要你帮我给 CO₂/Cu(111)/水/K⁺ 界面（188 原子，底部两层 32 个 Cu 固定）
> 做一套机器学习分子动力学：先 VASP 做 DFT 结构优化，再跑 VASP AIMD 出轨迹，
> 然后转成带能量和力的 MACE extxyz 数据集（按 100 帧时间块切 train/valid/test，**千万别随机打散**），
> 训一个 MACE 势函数（验收标准力误差 ≲10 meV/Å），最后导出成 LAMMPS 能加载的模型跑长时 MD——
> 目的就是把只能跑皮秒的 AIMD 推到纳秒，用来研究 CO₂ 的界面动力学；走**中性** AIMD。”
## 一句话 → 结构化规格
先把需求填进 `assets/campaign_spec.example.json` 并**直接按下列默认值自主落 `campaign_spec.json`，不向用户逐项确认科学参数**（泛函/色散/门限/温度/AIMD 时长/切分/超参一律取默认值，仅在正文标注为 `proposed_candidate`/`pending_verification`）；**唯一允许的用户交互是 S0 由 `onescience-runsite` 弹出的工作台连接选择**。本场景默认：
- `electrochemistry.mode = "neutral"`（禁 VASPsol/CP）。
- `slab.fix_bottom_layers = 2`（188 原子中底部 **32 个 Cu 固定**）。
- `dataset.split = {policy:"time-block", block_frames:100, shuffle:false}`（**绝不随机打散**）。
- `train.acceptance.forces_rmse_meV_per_A = 10.0`。
- `aimd = {NVT, Nose-Hoover, 300 K, 1 fs, 6 ps, velocity_init:"fresh-MB"}`。
- `export = {libtorch, elements:["H","C","O","K","Cu"]}`。
## 阶段、门限与交接（自包含）
…Read the whole file at its exact version.
How to install
mdr add onescience-ai/oneskills/onescience-mlmd-campaign-orchestrator@sha256:be02314327982969The line at the top of this page pins the label git:20261103.d90dac1, which is what most people want. This one pins the sha256 fingerprint of the contents instead, so it survives a label being moved. Either way the exact version is written to mdr.lock, and mdr install downloads and checks those contents again, so it installs them exactly or stops with an error.
[](https://markdownregistry.com/a/art_pcs5wo4jssn4zoue)
1 badge image fetches in 30 days. What the badge shows.
Versions
The registry has recorded 5 versions of this file, between Oct 15, 2026 and Nov 3, 2026, about one every 5 days.
| version | committed | commit | size | audit | |
|---|---|---|---|---|---|
| git:20261103.d90dac1 latest | 2026-11-03 | d90dac1 | 11,231 B | A | view · diff |
| git:20261030.3665286 | 2026-10-30 | 3665286 | 8,596 B | A | view · diff |
| git:20261019.b4be58a | 2026-10-19 | b4be58a | 7,777 B | A | view · diff |
| git:20261016.54d43b9 | 2026-10-16 | 54d43b9 | 7,250 B | A | view · diff |
| git:20261015.a4fa399 | 2026-10-15 | a4fa399 | 6,747 B | A | view |
Audit of the latest version
- pass: Frontmatter block present
- pass: Frontmatter declares a name
- pass: Frontmatter declares a description
- pass: Size between 200 bytes and 200 KB (11231 bytes)
- pass: No zero-width or bidi control characters
- pass: No instruction hidden inside an HTML comment
- pass: No link to an exfiltration or paste host
- pass: No credential-shaped string
- pass: No instruction to send local credentials anywhere
- pass: No text hidden with inline styles
- pass: No prompt-injection phrasing
- pass: No curl or wget piped into a shell
- pass: No recursive delete of root, home or parent
- pass: No instruction to read or print local credentials
- pass: No base64 blob over 200 characters
- pass: No link to a raw IP address
- pass: No script tag
Source
onescience-ai/oneskills · 24 stars · license MIT · pushed 2026-11-04 · branch master
API
GET https://markdownregistry.com/api/v1/artifacts/art_pcs5wo4jssn4zoue GET https://markdownregistry.com/api/v1/resolve?ref=onescience-ai/oneskills/onescience-mlmd-campaign-orchestrator GET https://markdownregistry.com/api/v1/blob/be02314327982969dd4cc00361d26398295dbab8bc6f7d5b142629606b6a94ef
Your AI assistant does the legwork. You hear about the deals worth your word. Hand yours the standing instructions at modelranch.com and it joins the network that reads files like this one.
Guides: the SKILL.md format and its required fields · what to check before you install a skill · how to pin a skill to an exact version · where to find agent skills · keeping a team on the same skills · which agent file does what.