chore(skills): import DeePMD-kit agent skills#5437
Merged
Conversation
Contributor
📝 WalkthroughWalkthroughThis pull request adds a comprehensive skill documentation library for the DeePMD-kit molecular dynamics and machine learning toolkit. Seven skill documents and supporting assets are introduced, covering model training (DPA3 and SE_E2_A), fine-tuning, Python inference, and LAMMPS integration, enabling AI agents to execute complete workflows end-to-end. ChangesDeePMD-kit Skill Documentation Library
Estimated code review effort🎯 1 (Trivial) | ⏱️ ~5 minutes 🚥 Pre-merge checks | ✅ 5✅ Passed checks (5 passed)
✏️ Tip: You can configure your own custom pre-merge checks in the settings. ✨ Finishing Touches🧪 Generate unit tests (beta)
Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out. Comment |
![coderabbitai[bot]](https://avatars.githubusercontent.com/in/347564?s=60&v=4){kind=small}
Contributor
There was a problem hiding this comment.
Actionable comments posted: 3
🧹 Nitpick comments (1)
.github/skills/lammps-deepmd/references/commands-and-workflow.md (1)
9-11: 💤 Low valueConsider varying sentence structure for readability.
Three consecutive items begin with "If the user," which slightly reduces readability. Consider rephrasing one to improve flow.
♻️ Optional rephrasing
1. If the user asks to run a simulation, always confirm the structure file and DeePMD model file before execution. -1. If the user asks for offline execution, ask which exact LAMMPS command should be used instead of guessing. +1. For offline execution requests, ask which exact LAMMPS command should be used instead of guessing. 1. If the user only asks for a template, do not overcomplicate it with advanced computes or fixes unless they are needed.🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In @.github/skills/lammps-deepmd/references/commands-and-workflow.md around lines 9 - 11, The three checklist items all start with "If the user..." which hurts flow; edit the three bullets (the lines beginning "If the user asks to run a simulation", "If the user asks for offline execution", and "If the user only asks for a template") so at least one uses a different sentence structure — for example: "Confirm the structure file and DeePMD model file before running a simulation" (replace the first), "For offline execution, ask which exact LAMMPS command should be used" (rephrase the second), or "Provide a simple template unless advanced computes or fixes are requested" (rephrase the third) — keep meaning identical and adjust punctuation/ordering for variety and readability.
🤖 Prompt for all review comments with AI agents
Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.
Inline comments:
In @.github/skills/deepmd-python-inference/SKILL.md:
- Around line 189-195: The fenced code block showing the "dp test" output is
missing a language tag causing markdownlint MD040; update the triple-backtick
fence that encloses the Energy/Force/Virial lines to include a language tag
(e.g., text) so it becomes ```text at the start of the block while leaving the
contents unchanged.
In @.github/skills/deepmd-train-dpa3/SKILL.md:
- Around line 42-52: The markdown has two fenced code blocks missing language
identifiers (MD040): the directory tree block starting with "system_dir/" and
the lcurve header block starting with "# step rmse_val rmse_trn rmse_e_val
..."; update those fences in SKILL.md by adding a language label (e.g., ```text)
to both blocks (including the similar block at lines ~201-203) so the opening
triple-backticks become ```text for each affected code block.
In @.github/skills/deepmd-train-se-e2-a/SKILL.md:
- Around line 42-51: The Markdown has unlabeled fenced code blocks that trigger
MD040; update the two plain ``` fences in SKILL.md to include a language tag
(e.g., change ``` to ```text) for the directory-tree block (the block showing
"system_dir/ ... box.npy") and the tabular/header block (the line starting with
"# step rmse_val..."); also update the additional occurrence around lines
167-169 to use ```text so all three fenced blocks are labeled.
---
Nitpick comments:
In @.github/skills/lammps-deepmd/references/commands-and-workflow.md:
- Around line 9-11: The three checklist items all start with "If the user..."
which hurts flow; edit the three bullets (the lines beginning "If the user asks
to run a simulation", "If the user asks for offline execution", and "If the user
only asks for a template") so at least one uses a different sentence structure —
for example: "Confirm the structure file and DeePMD model file before running a
simulation" (replace the first), "For offline execution, ask which exact LAMMPS
command should be used" (rephrase the second), or "Provide a simple template
unless advanced computes or fixes are requested" (rephrase the third) — keep
meaning identical and adjust punctuation/ordering for variety and readability.
🪄 Autofix (Beta)
Fix all unresolved CodeRabbit comments on this PR:
- Push a commit to this branch (recommended)
- Create a new PR with the fixes
ℹ️ Review info
⚙️ Run configuration
Configuration used: Repository UI
Review profile: CHILL
Plan: Pro
Run ID: 60ee211d-e03e-4446-8bc5-04eaca547492
📒 Files selected for processing (7)
.github/skills/deepmd-finetune-dpa3/SKILL.md.github/skills/deepmd-python-inference/SKILL.md.github/skills/deepmd-train-dpa3/SKILL.md.github/skills/deepmd-train-se-e2-a/SKILL.md.github/skills/lammps-deepmd/SKILL.md.github/skills/lammps-deepmd/assets/input.nvt.lammps.github/skills/lammps-deepmd/references/commands-and-workflow.md
Comment on lines
+189
to
+195
| ``` | ||
| Energy RMSE : 1.234e-03 eV | ||
| Energy RMSE/Natoms : 6.427e-06 eV | ||
| Force RMSE : 2.345e-02 eV/A | ||
| Virial RMSE : 5.678e-02 eV | ||
| Virial RMSE/Natoms : 2.957e-04 eV | ||
| ``` |
Contributor
There was a problem hiding this comment.
Add a language tag to the fenced output block.
The code fence for dp test output is unlabeled; markdownlint MD040 will flag it.
Suggested fix
-```
+```text
Energy RMSE : 1.234e-03 eV
Energy RMSE/Natoms : 6.427e-06 eV
Force RMSE : 2.345e-02 eV/A
Virial RMSE : 5.678e-02 eV
Virial RMSE/Natoms : 2.957e-04 eV</details>
<!-- suggestion_start -->
<details>
<summary>📝 Committable suggestion</summary>
> ‼️ **IMPORTANT**
> Carefully review the code before committing. Ensure that it accurately replaces the highlighted code, contains no missing lines, and has no issues with indentation. Thoroughly test & benchmark the code to ensure it meets the requirements.
```suggestion
🧰 Tools
🪛 markdownlint-cli2 (0.22.1)
[warning] 189-189: Fenced code blocks should have a language specified
(MD040, fenced-code-language)
🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.
In @.github/skills/deepmd-python-inference/SKILL.md around lines 189 - 195, The
fenced code block showing the "dp test" output is missing a language tag causing
markdownlint MD040; update the triple-backtick fence that encloses the
Energy/Force/Virial lines to include a language tag (e.g., text) so it becomes
```text at the start of the block while leaving the contents unchanged.
Comment on lines
+42
to
+52
| ``` | ||
| system_dir/ | ||
| ├── type.raw | ||
| ├── type_map.raw | ||
| └── set.000/ | ||
| ├── coord.npy | ||
| ├── energy.npy | ||
| ├── force.npy | ||
| ├── box.npy | ||
| └── virial.npy | ||
| ``` |
Contributor
There was a problem hiding this comment.
Label fenced blocks to satisfy markdownlint.
Two fenced blocks are missing language identifiers (directory tree and lcurve header), which triggers MD040.
Suggested fix
-```
+```text
system_dir/
├── type.raw
├── type_map.raw
└── set.000/
├── coord.npy
├── energy.npy
├── force.npy
├── box.npy
└── virial.npy...
- +text
step rmse_val rmse_trn rmse_e_val rmse_e_trn rmse_f_val rmse_f_trn rmse_v_val rmse_v_trn lr
Also applies to: 201-203
🧰 Tools
🪛 markdownlint-cli2 (0.22.1)
[warning] 42-42: Fenced code blocks should have a language specified
(MD040, fenced-code-language)
🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.
In @.github/skills/deepmd-train-dpa3/SKILL.md around lines 42 - 52, The markdown
has two fenced code blocks missing language identifiers (MD040): the directory
tree block starting with "system_dir/" and the lcurve header block starting with
"# step rmse_val rmse_trn rmse_e_val ..."; update those fences in SKILL.md
by adding a language label (e.g., ```text) to both blocks (including the similar
block at lines ~201-203) so the opening triple-backticks become ```text for each
affected code block.
Comment on lines
+42
to
+51
| ``` | ||
| system_dir/ | ||
| ├── type.raw # atom type indices, one integer per atom | ||
| ├── type_map.raw # element names, one per line | ||
| └── set.000/ | ||
| ├── coord.npy # coordinates (nframes, natoms*3) | ||
| ├── energy.npy # energies (nframes, 1) | ||
| ├── force.npy # forces (nframes, natoms*3) | ||
| └── box.npy # cell vectors (nframes, 9) | ||
| ``` |
Contributor
There was a problem hiding this comment.
Add language tags to unlabeled fenced blocks.
The two plain fenced blocks should be tagged (e.g., text) to avoid MD040 lint warnings.
Suggested fix
-```
+```text
system_dir/
├── type.raw # atom type indices, one integer per atom
├── type_map.raw # element names, one per line
└── set.000/
├── coord.npy # coordinates (nframes, natoms*3)
├── energy.npy # energies (nframes, 1)
├── force.npy # forces (nframes, natoms*3)
└── box.npy # cell vectors (nframes, 9)...
- +text
step rmse_val rmse_trn rmse_e_val rmse_e_trn rmse_f_val rmse_f_trn rmse_v_val rmse_v_trn lr
Also applies to: 167-169
🧰 Tools
🪛 markdownlint-cli2 (0.22.1)
[warning] 42-42: Fenced code blocks should have a language specified
(MD040, fenced-code-language)
🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.
In @.github/skills/deepmd-train-se-e2-a/SKILL.md around lines 42 - 51, The
Markdown has unlabeled fenced code blocks that trigger MD040; update the two
plain ``` fences in SKILL.md to include a language tag (e.g., change ``` to
```text) for the directory-tree block (the block showing "system_dir/ ...
box.npy") and the tabular/header block (the line starting with "# step
rmse_val..."); also update the additional occurrence around lines 167-169 to use
```text so all three fenced blocks are labeled.
Codecov Report✅ All modified and coverable lines are covered by tests. Additional details and impacted files@@ Coverage Diff @@
## master #5437 +/- ##
==========================================
- Coverage 82.47% 82.46% -0.01%
==========================================
Files 825 825
Lines 87721 87722 +1
Branches 4206 4206
==========================================
Hits 72344 72344
Misses 14094 14094
- Partials 1283 1284 +1 ☔ View full report in Codecov by Sentry. 🚀 New features to boost your workflow:
|
Imported from jinzhezenggroup/computational-chemistry-agent-skills. Upstream-Commit: jinzhezenggroup/computational-chemistry-agent-skills@155a675 Upstream-Paths: - machine-learning-potentials/deepmd-finetune-dpa3 - machine-learning-potentials/deepmd-python-inference - machine-learning-potentials/deepmd-train-dpa3 - machine-learning-potentials/deepmd-train-se-e2-a - molecular-dynamics/lammps-deepmd
Imported from jinzhezenggroup/computational-chemistry-agent-skills. Upstream-Commit: jinzhezenggroup/computational-chemistry-agent-skills@d8ff524 Upstream-Paths: - machine-learning-potentials/deepmd-finetune-dpa3 - machine-learning-potentials/deepmd-python-inference - machine-learning-potentials/deepmd-train-dpa3 - machine-learning-potentials/deepmd-train-se-e2-a - molecular-dynamics/lammps-deepmd
…deepmodeling#30) Imported from jinzhezenggroup/computational-chemistry-agent-skills. Upstream-Commit: jinzhezenggroup/computational-chemistry-agent-skills@3b9a0df Upstream-Paths: - machine-learning-potentials/deepmd-finetune-dpa3 - machine-learning-potentials/deepmd-python-inference - machine-learning-potentials/deepmd-train-dpa3 - machine-learning-potentials/deepmd-train-se-e2-a - molecular-dynamics/lammps-deepmd
…ng#37) Imported from jinzhezenggroup/computational-chemistry-agent-skills. Upstream-Commit: jinzhezenggroup/computational-chemistry-agent-skills@0fbe96d Upstream-Paths: - machine-learning-potentials/deepmd-finetune-dpa3 - machine-learning-potentials/deepmd-python-inference - machine-learning-potentials/deepmd-train-dpa3 - machine-learning-potentials/deepmd-train-se-e2-a - molecular-dynamics/lammps-deepmd
…modeling#35) Imported from jinzhezenggroup/computational-chemistry-agent-skills. Upstream-Commit: jinzhezenggroup/computational-chemistry-agent-skills@0a9651d Upstream-Paths: - machine-learning-potentials/deepmd-finetune-dpa3 - machine-learning-potentials/deepmd-python-inference - machine-learning-potentials/deepmd-train-dpa3 - machine-learning-potentials/deepmd-train-se-e2-a - molecular-dynamics/lammps-deepmd
…pmodeling#52) Imported from jinzhezenggroup/computational-chemistry-agent-skills. Upstream-Commit: jinzhezenggroup/computational-chemistry-agent-skills@622da72 Upstream-Paths: - machine-learning-potentials/deepmd-finetune-dpa3 - machine-learning-potentials/deepmd-python-inference - machine-learning-potentials/deepmd-train-dpa3 - machine-learning-potentials/deepmd-train-se-e2-a - molecular-dynamics/lammps-deepmd
Imported from jinzhezenggroup/computational-chemistry-agent-skills. Upstream-Commit: jinzhezenggroup/computational-chemistry-agent-skills@3f03f72 Upstream-Paths: - machine-learning-potentials/deepmd-finetune-dpa3 - machine-learning-potentials/deepmd-python-inference - machine-learning-potentials/deepmd-train-dpa3 - machine-learning-potentials/deepmd-train-se-e2-a - molecular-dynamics/lammps-deepmd
Imported from jinzhezenggroup/computational-chemistry-agent-skills. Upstream-Commit: jinzhezenggroup/computational-chemistry-agent-skills@37ed5a2 Upstream-Paths: - machine-learning-potentials/deepmd-finetune-dpa3 - machine-learning-potentials/deepmd-python-inference - machine-learning-potentials/deepmd-train-dpa3 - machine-learning-potentials/deepmd-train-se-e2-a - molecular-dynamics/lammps-deepmd
njzjz-bot
force-pushed
the
chore/import-agent-skills
branch
from
39ba8b9 to
5156c40
Compare
Contributor
There was a problem hiding this comment.
♻️ Duplicate comments (3)
skills/deepmd-train-se-e2-a/SKILL.md (1)
42-51:⚠️ Potential issue | 🟡 Minor | ⚡ Quick winAdd
textlanguage tags to unlabeled fences (MD040).At Line 42 and Line 167, plain triple-backtick blocks are missing a language identifier.
Suggested fix
-``` +```text system_dir/ ├── type.raw # atom type indices, one integer per atom ├── type_map.raw # element names, one per line └── set.000/ ├── coord.npy # coordinates (nframes, natoms*3) ├── energy.npy # energies (nframes, 1) ├── force.npy # forces (nframes, natoms*3) └── box.npy # cell vectors (nframes, 9)...
-
+textstep rmse_val rmse_trn rmse_e_val rmse_e_trn rmse_f_val rmse_f_trn rmse_v_val rmse_v_trn lr
Also applies to: 167-169
🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@skills/deepmd-train-se-e2-a/SKILL.md` around lines 42 - 51, The Markdown file SKILL.md contains unlabeled triple-backtick code fences (the directory-tree block starting with "system_dir/" and the stats/header block starting with "# step rmse_val...") which trigger MD040; update those fences to include a language tag (e.g., change ``` to ```text) so the blocks are labeled as plain text; locate the unlabeled fences by searching for the literal lines beginning with "system_dir/" and the stats header line "# step rmse_val" and replace their opening fences with ```text (and keep the closing ``` unchanged).skills/deepmd-python-inference/SKILL.md (1)
189-195:⚠️ Potential issue | 🟡 Minor | ⚡ Quick winLabel the
dp testoutput fence astext(MD040).At Line 189, the fenced block is unlabeled and will trigger markdownlint.
Suggested fix
-``` +```text Energy RMSE : 1.234e-03 eV Energy RMSE/Natoms : 6.427e-06 eV Force RMSE : 2.345e-02 eV/A Virial RMSE : 5.678e-02 eV Virial RMSE/Natoms : 2.957e-04 eV</details> <details> <summary>🤖 Prompt for AI Agents</summary>Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.In
@skills/deepmd-python-inference/SKILL.mdaround lines 189 - 195, The fenced
output block showing the "dp test" results is unlabeled and triggers
markdownlint MD040; update that fenced code block (the dp test output block
containing "Energy RMSE ... Virial RMSE/Natoms") by adding the language/label
"text" immediately after the opening triple backticks (i.e. change ``` toskills/deepmd-train-dpa3/SKILL.md (1)
42-52:⚠️ Potential issue | 🟡 Minor | ⚡ Quick winAdd language tags to plain fenced blocks (MD040).
At Line 42 and Line 201, the opening fences are unlabeled. Please mark them as
textto satisfy markdownlint consistently.Suggested fix
-``` +```text system_dir/ ├── type.raw ├── type_map.raw └── set.000/ ├── coord.npy ├── energy.npy ├── force.npy ├── box.npy └── virial.npy...
-
+textstep rmse_val rmse_trn rmse_e_val rmse_e_trn rmse_f_val rmse_f_trn rmse_v_val rmse_v_trn lr
Also applies to: 201-203
🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@skills/deepmd-train-dpa3/SKILL.md` around lines 42 - 52, In SKILL.md, the two unlabeled fenced code blocks starting with "system_dir/" and the table header line starting with "# step rmse_val..." should have a language tag; change their opening fences from ``` to ```text so both code fences are labeled (i.e., replace the first ``` before the block that begins with "system_dir/" and the ``` before the block that begins with "# step rmse_val rmse_trn..." with ```text).
🤖 Prompt for all review comments with AI agents
Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.
Duplicate comments:
In `@skills/deepmd-python-inference/SKILL.md`:
- Around line 189-195: The fenced output block showing the "dp test" results is
unlabeled and triggers markdownlint MD040; update that fenced code block (the dp
test output block containing "Energy RMSE ... Virial RMSE/Natoms") by adding the
language/label "text" immediately after the opening triple backticks (i.e.
change ``` to ```text) so the block is explicitly labeled as text.
In `@skills/deepmd-train-dpa3/SKILL.md`:
- Around line 42-52: In SKILL.md, the two unlabeled fenced code blocks starting
with "system_dir/" and the table header line starting with "# step
rmse_val..." should have a language tag; change their opening fences from ``` to
```text so both code fences are labeled (i.e., replace the first ``` before the
block that begins with "system_dir/" and the ``` before the block that begins
with "# step rmse_val rmse_trn..." with ```text).
In `@skills/deepmd-train-se-e2-a/SKILL.md`:
- Around line 42-51: The Markdown file SKILL.md contains unlabeled
triple-backtick code fences (the directory-tree block starting with
"system_dir/" and the stats/header block starting with "# step rmse_val...")
which trigger MD040; update those fences to include a language tag (e.g., change
``` to ```text) so the blocks are labeled as plain text; locate the unlabeled
fences by searching for the literal lines beginning with "system_dir/" and the
stats header line "# step rmse_val" and replace their opening fences with
```text (and keep the closing ``` unchanged).
ℹ️ Review info
⚙️ Run configuration
Configuration used: Repository UI
Review profile: CHILL
Plan: Pro
Run ID: f6179bb2-babd-4ded-8a1e-47f5c7f392f3
📒 Files selected for processing (7)
skills/deepmd-finetune-dpa3/SKILL.mdskills/deepmd-python-inference/SKILL.mdskills/deepmd-train-dpa3/SKILL.mdskills/deepmd-train-se-e2-a/SKILL.mdskills/lammps-deepmd/SKILL.mdskills/lammps-deepmd/assets/input.nvt.lammpsskills/lammps-deepmd/references/commands-and-workflow.md
✅ Files skipped from review due to trivial changes (2)
- skills/lammps-deepmd/assets/input.nvt.lammps
- skills/lammps-deepmd/SKILL.md
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
4 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Problem
Change
skills/.Notes
machine-learning-potentials/deepmd-*,molecular-dynamics/lammps-deepmd.skills/deepmd-*,skills/lammps-deepmd.git diff --check origin/master...HEAD.Authored by OpenClaw (model: gpt-5.5)
Summary by CodeRabbit