onescience-dft-model-gate SKILL.md skillA
onescience-dft-model-gate 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-dft-model-gate@git:20261030.3665286Pins git:20261030.3665286 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-dft-model-gate/SKILL.md- Versions recorded
- 7
- Latest version
- git:20261030.3665286, committed Oct 30, 2026
- First version recorded
- Oct 14, 2026
- Size
- 11,050 bytes
- Repository stars
- 24
- License
- MIT
DFT/VASP 初始原子结构建模与交付门限。任一信号命中必用：(a) 表面/界面：Cu/Fe/Pt/Pd/Ni/Au/Ag/TiO2/MoS2/石墨烯/ZnO/CeO2 等 slab、(111)/(100)/(110)/(001) 晶面、异质结、界面建模；(b) 吸附/溶剂化：吸附物放置、CO2/CO/H2O/N2/OH/H/O 等分子加到表面、显式水/水层/水膜、电解质、溶剂化；(c) 掺杂/缺陷：K/Li/Na/Ca 掺杂、空位、取代、表面重构；(d) VASP 输入交付：POSCAR/CONTCAR/INCAR/KPOINTS/POTCAR、model_contract.json、selective dynamics、固定底部原子、ENCUT、赝势、几何合理性；(e) 场景关键词：初始模型、起始结构、建模交付、计算前验收、VASP 计算前。构建前必先写 model_contract.json，通过硬性验收门限才能提交。本技能为 executor 类型，通常由 onescience-orchestrator 调度；用户未点名时优先走 orchestrator 入口。.
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
# DFT 模型就绪门限（onescience-dft-model-gate）
## 这个技能解决什么
用基座能力"手搓"一个原子结构并写一份说明，**在几何上看起来没问题、实际不可计算**，是这类任务的
高频失效模式。典型症状（都真实发生过）：
| 症状 | 本质 |
|---|---|
| 两个 H 相距 1.0 Å | 分子取向按"最近邻"硬投，互相为最近邻的分子把 H 迎面对撞 |
| 某个 O 接受 5 个氢键 | 没有全局供体-受体匹配，受体被过度配位 |
| 交付说明写"最近非键距离 ≥ 2.36 Å"，文件实测 1.01 Å | **校验数字不是在这份文件上跑出来的** |
| 吸附物 2.5 Å 落在金属原子上方，却标称"物理吸附" | 声明意图与几何矛盾，无人核对 |
| 水膜标称"非晶态/近液态"，实为两层完全共面 | 无维度/密度口径，靠形容词 |
| KPOINTS 多写一行注释 | VASP 第 2 行必须是 Nk，多余注释会让作业直接读失败 |
| INCAR 注释写"K_pv ~ 250 eV"，实际 116.7 eV | 注释是不可信的二手信息，必须从 POTCAR 现算 |
本技能把"交付前必须自证"变成**可执行的硬门限**，并且**物理门限与体系无关**（全部由元素半径数据推导），
体系相关的预期由一份声明式契约给出。换任何材料/分子/溶剂都能用，不需要改检查器。
## 强制流程
### 1. 先写 `model_contract.json`（建模之前，不是之后）
契约是"这个模型的物理预期"。检查器只核对几何是否满足你的声明，不替你猜意图。
完整字段见 `assets/model_contract.example.json`。最小可用形态：
```json
{
"system": "<吸附物> on <载体>(hkl) with <溶剂>/<离子>",
"output_species_order": ["H", "C", "O", "K", "Cu"],
"delivery_form": "A",
"framework": { "elements": ["Cu"] },
"species": {
"H2O": { "count": 40, "role": "solvent", "solvent_model": "liquid-film", "layers_min": 3 },
"CO2": { "count": 1, "role": "adsorbate", "mode": "physisorbed" },
"K": { "count": 1, "role": "ion" }
},
…Read the whole file at its exact version.
How to install
mdr add onescience-ai/oneskills/onescience-dft-model-gate@sha256:071d71471583e822The line at the top of this page pins the label git:20261030.3665286, 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_hjrlyyb3mbz33h2v)
4 badge image fetches in 30 days. What the badge shows.
Versions
The registry has recorded 7 versions of this file, between Oct 14, 2026 and Oct 30, 2026, about one every 3 days.
| version | committed | commit | size | audit | |
|---|---|---|---|---|---|
| git:20261030.3665286 latest | 2026-10-30 | 3665286 | 11,050 B | A | view · diff |
| git:20261021.773153b | 2026-10-21 | 773153b | 10,210 B | A | view · diff |
| git:20261018.30e823c | 2026-10-18 | 30e823c | 9,742 B | A | view · diff |
| git:20261016.3f022d7 | 2026-10-16 | 3f022d7 | 9,001 B | A | view · diff |
| git:20261016.364592c | 2026-10-16 | 364592c | 8,877 B | A | view · diff |
| git:20261015.b018fe9 | 2026-10-15 | b018fe9 | 8,269 B | A | view · diff |
| git:20261014.a61dc0a | 2026-10-14 | a61dc0a | 6,711 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 (11050 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_hjrlyyb3mbz33h2v GET https://markdownregistry.com/api/v1/resolve?ref=onescience-ai/oneskills/onescience-dft-model-gate GET https://markdownregistry.com/api/v1/blob/071d71471583e8222f5cc3a3405b985a7f62b29eb6e660f44121df085f9b0566
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.