docs(agent-skills): add DeePMD-kit skill guide#5497
Conversation
njzjz-bot
force-pushed
the
docs-agent-skills
branch
from
618e32e to
eb52354
Compare
|
I accidentally created the initial head branch from an incomplete tree, which made the PR temporarily show mass deletions. I have force-corrected the PR head to a clean commit based on current upstream Current diff is now limited to:
Verified with Sorry for the noise. Authored by OpenClaw (model: custom-chat-jinzhezeng-group/gpt-5.5) |
|
Note Reviews pausedIt looks like this branch is under active development. To avoid overwhelming you with review comments due to an influx of new commits, CodeRabbit has automatically paused this review. You can configure this behavior by changing the Use the following commands to manage reviews:
Use the checkboxes below for quick actions:
📝 WalkthroughWalkthroughAdds ChangesAgent Skills Documentation
Estimated code review effort🎯 1 (Trivial) | ⏱️ ~3 minutes Possibly related PRs
Suggested labels
Suggested reviewers
🚥 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 |
njzjz-bot
force-pushed
the
docs-agent-skills
branch
from
88c0ebd to
7b92f72
Compare
![coderabbitai[bot]](https://avatars.githubusercontent.com/in/347564?s=60&v=4){kind=small}
There was a problem hiding this comment.
🧹 Nitpick comments (2)
doc/agent-skills.md (2)
23-23: ⚡ Quick winStandardize “pretrained” terminology in the same sentence.
Line 23 mixes “pre-trained” and “pretrained” style in nearby phrasing; use one form consistently to reduce wording drift in docs.
🤖 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 `@doc/agent-skills.md` at line 23, The sentence in doc/agent-skills.md mixes “pre-trained” and “pretrained”; standardize to a single spelling (e.g., change “pre-trained” to “pretrained” or vice versa) so the phrase "multi-task pre-trained models, or built-in models downloaded by `dp pretrained download`" uses the same form for both occurrences of the term in that sentence.
62-62: ⚡ Quick winAlign dialect choice for “afterward(s)” with the docs style.
Line 62 uses “afterwards”; if this repo standard is American English, switch to “afterward” for consistency.
🤖 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 `@doc/agent-skills.md` at line 62, Replace the British/variant "afterwards" in the sentence "Refresh or restart the session afterwards so the installed skills are reloaded." with the American English "afterward" to match the repo's style; update that sentence to read "Refresh or restart the session afterward so the installed skills are reloaded." and confirm any other occurrences of "afterwards" in doc/agent-skills.md are also changed to "afterward".
🤖 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.
Nitpick comments:
In `@doc/agent-skills.md`:
- Line 23: The sentence in doc/agent-skills.md mixes “pre-trained” and
“pretrained”; standardize to a single spelling (e.g., change “pre-trained” to
“pretrained” or vice versa) so the phrase "multi-task pre-trained models, or
built-in models downloaded by `dp pretrained download`" uses the same form for
both occurrences of the term in that sentence.
- Line 62: Replace the British/variant "afterwards" in the sentence "Refresh or
restart the session afterwards so the installed skills are reloaded." with the
American English "afterward" to match the repo's style; update that sentence to
read "Refresh or restart the session afterward so the installed skills are
reloaded." and confirm any other occurrences of "afterwards" in
doc/agent-skills.md are also changed to "afterward".
ℹ️ Review info
⚙️ Run configuration
Configuration used: Repository UI
Review profile: CHILL
Plan: Pro
Run ID: b81746c2-045e-4018-b0d1-d6240174bd9f
📒 Files selected for processing (2)
doc/agent-skills.mddoc/index.rst
✅ Files skipped from review due to trivial changes (1)
- doc/index.rst
njzjz-bot
force-pushed
the
docs-agent-skills
branch
2 times, most recently
from
0857316 to
ef48db0
Compare
There was a problem hiding this comment.
Actionable comments posted: 1
🤖 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 `@doc/agent-skills.md`:
- Around line 21-23: Summary: Inconsistent use of “pretrained” vs “pre-trained”
in prose. Fix: standardize prose to a single variant (choose either "pretrained"
or "pre-trained") across the sentence that references the CLI literal `dp
pretrained download` and the skill names `deepmd-finetune-dpa3` /
`deepmd-train-se-e2-a`, but do not change the command literal itself; update the
wording on the line describing `deepmd-finetune-dpa3` so it matches the chosen
prose form used elsewhere in doc/agent-skills.md.
🪄 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: c1dd5dea-6e61-44b5-a20e-ad7043072f71
📒 Files selected for processing (2)
doc/agent-skills.mddoc/index.rst
🚧 Files skipped from review as they are similar to previous changes (1)
- doc/index.rst
Use concrete OpenClaw install examples instead of shell-hostile angle-bracket placeholders, and explain how to change the target agent. Authored by OpenClaw (model: custom-chat-jinzhezeng-group/gpt-5.5)
njzjz-bot
force-pushed
the
docs-agent-skills
branch
from
3995222 to
c35e1a5
Compare
for more information, see https://pre-commit.ci
There was a problem hiding this comment.
Actionable comments posted: 1
🧹 Nitpick comments (1)
doc/agent-skills.md (1)
63-64: ⚡ Quick winPin the GitHub install source to a tag or commit for reproducibility.
Using
.../tree/master/skillsmakes installs drift asmasterchanges. For stable docs behavior, prefer a release tag or commit SHA in the example command.Suggested doc edit
-npx -y skills add https://github.com/deepmodeling/deepmd-kit/tree/master/skills \ +npx -y skills add https://github.com/deepmodeling/deepmd-kit/tree/<tag-or-commit>/skills \ -a openclaw -y🤖 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 `@doc/agent-skills.md` around lines 63 - 64, The example install command currently points at the floating branch "tree/master/skills" (npx -y skills add https://github.com/deepmodeling/deepmd-kit/tree/master/skills -a openclaw -y); update that URL to pin to a specific release tag or commit SHA (e.g. replace /tree/master/skills with /tree/<release-tag>/skills or /commit/<commit-sha>/skills) so the example is reproducible and won’t drift as master changes.
🤖 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 `@doc/agent-skills.md`:
- Line 71: In the sentence containing "Refresh or restart the session afterwards
so the installed skills", replace the US-English variant by changing
"afterwards" to "afterward" so the line reads "Refresh or restart the session
afterward so the installed skills"; update that exact phrase in agent-skills.md
to maintain consistency across the docs.
---
Nitpick comments:
In `@doc/agent-skills.md`:
- Around line 63-64: The example install command currently points at the
floating branch "tree/master/skills" (npx -y skills add
https://github.com/deepmodeling/deepmd-kit/tree/master/skills -a openclaw -y);
update that URL to pin to a specific release tag or commit SHA (e.g. replace
/tree/master/skills with /tree/<release-tag>/skills or
/commit/<commit-sha>/skills) so the example is reproducible and won’t drift as
master changes.
🪄 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: 1e71c387-16e7-489f-92a4-c38c56b8d707
📒 Files selected for processing (2)
doc/agent-skills.mddoc/index.rst
✅ Files skipped from review due to trivial changes (1)
- doc/index.rst
There was a problem hiding this comment.
Pull request overview
Adds a dedicated documentation page describing DeePMD-kit’s Agent Skills (including installation and lightweight verification guidance) and exposes it in the Sphinx navigation under the “Advanced” section.
Changes:
- Add new
doc/agent-skills.mdpage describing available DeePMD-kit skills and a related arXiv reference. - Document skill installation commands (local checkout vs GitHub URL) and minimal verification prompts.
- Link the new page from
doc/index.rstso it appears in the rendered docs.
Reviewed changes
Copilot reviewed 2 out of 2 changed files in this pull request and generated 3 comments.
| File | Description |
|---|---|
| doc/index.rst | Adds the new Agent Skills page to the “Advanced” toctree. |
| doc/agent-skills.md | New documentation page describing skills, installation, and verification guidance. |
💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.
Apply review feedback by keeping Markdown link text together, documenting the Node.js/npm prerequisite, normalizing terminology and US-English wording, and pinning the GitHub install example to a commit for reproducibility. Authored by OpenClaw (model: custom-chat-jinzhezeng-group/gpt-5.5)
for more information, see https://pre-commit.ci
Keep the remote installation example tracking the maintained master branch instead of pinning it to a commit, so users install the latest skills maintained in the main repository. Authored by OpenClaw (model: custom-chat-jinzhezeng-group/gpt-5.5)
for more information, see https://pre-commit.ci
|
Addressed the active review threads:
I did not keep the CodeRabbit suggestion to pin the GitHub install URL to a commit. This page documents skills maintained directly in the main DeePMD-kit repository, so the remote install example should track Authored by OpenClaw (version: 2026.5.28, model: custom-chat-jinzhezeng-group/gpt-5.5) |
Codecov Report✅ All modified and coverable lines are covered by tests. Additional details and impacted files@@ Coverage Diff @@
## master #5497 +/- ##
=======================================
Coverage 81.37% 81.38%
=======================================
Files 868 868
Lines 96598 96599 +1
Branches 4242 4242
=======================================
+ Hits 78611 78614 +3
+ Misses 16684 16681 -3
- Partials 1303 1304 +1 ☔ View full report in Codecov by Harness. 🚀 New features to boost your workflow:
|
Problem
Change
doc/agent-skills.mdpage modeled after the DPDispatcher Agent Skills page.jinzhezenggroup/computational-chemistry-agent-skills.Notes
jinzhezenggroup/computational-chemistry-agent-skillsat local HEAD81543b8after fetching.markdownlint-cli2 doc/agent-skills.mdpasses.uvfailed while downloading build requirements (numpy==2.4.6) due to a connection reset before the docs build could start.Authored by OpenClaw (model: custom-chat-jinzhezeng-group/gpt-5.5)
Summary by CodeRabbit