docs(skills): consolidate training skill disclosure#5530
Queued
njzjz-bot wants to merge 1 commit into
Queued
Conversation
Contributor
|
No actionable comments were generated in the recent review. 🎉 ℹ️ Recent review info⚙️ Run configurationConfiguration used: Repository UI Review profile: CHILL Plan: Pro Run ID: 📒 Files selected for processing (6)
💤 Files with no reviewable changes (2)
✅ Files skipped from review due to trivial changes (3)
📝 WalkthroughWalkthroughThis PR consolidates DeePMD-kit training documentation by replacing two separate model-specific skills ( ChangesDeePMD Training Skills Consolidation
Estimated code review effort🎯 2 (Simple) | ⏱️ ~12 minutes Possibly related PRs
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 |
![coderabbitai[bot]](https://avatars.githubusercontent.com/in/347564?s=60&v=4){kind=small}
Contributor
There was a problem hiding this comment.
🧹 Nitpick comments (1)
skills/deepmd-train/SKILL.md (1)
22-28: 💤 Low valueConsider rewording successive "If" clauses to improve readability.
Lines 26–28 each begin with "If", which may reduce clarity. Rewording some conditionals to vary sentence structure would improve the flow:
- Do not start by reading every model document. First classify the request: - - - If the user already named a model, read only that model reference. - - If the user asks for a recommendation, collect the decision inputs below, choose a model, then read only the selected reference. - - If model-specific parameters are not needed yet, stay in this top-level workflow. + Do not start by reading every model document. First classify the request: + + - When the user names a model, read only that model reference. + - For recommendation requests, collect the decision inputs below, choose a model, then read only the selected reference. + - When model-specific parameters are not needed yet, stay in this top-level workflow.🤖 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/SKILL.md` around lines 22 - 28, The three consecutive conditional list items under "Progressive disclosure protocol" start each with "If", which reduces readability; edit the three bullets so they vary sentence openings (e.g., keep the first as "If the user already named a model,...", change the second to "When the user asks for a recommendation, collect the decision inputs..., then read only the selected reference", and change the third to "If model-specific parameters are not needed yet, remain in the top-level workflow" or use lead-ins like "Otherwise," or "In that case,"), preserving meaning and the existing decision flow in the SKILL.md section.Source: Linters/SAST tools
🤖 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 `@skills/deepmd-train/SKILL.md`:
- Around line 22-28: The three consecutive conditional list items under
"Progressive disclosure protocol" start each with "If", which reduces
readability; edit the three bullets so they vary sentence openings (e.g., keep
the first as "If the user already named a model,...", change the second to "When
the user asks for a recommendation, collect the decision inputs..., then read
only the selected reference", and change the third to "If model-specific
parameters are not needed yet, remain in the top-level workflow" or use lead-ins
like "Otherwise," or "In that case,"), preserving meaning and the existing
decision flow in the SKILL.md section.
ℹ️ Review info
⚙️ Run configuration
Configuration used: Repository UI
Review profile: CHILL
Plan: Pro
Run ID: ec4dfae2-8b0d-4442-8ec2-1a0455dcfad6
📒 Files selected for processing (5)
doc/development/agent-skills.mddoc/index.rstskills/deepmd-train/SKILL.mdskills/deepmd-train/models/dpa3.mdskills/deepmd-train/models/se-e2-a.md
njzjz
force-pushed
the
skill-progressive-train
branch
from
51e345e to
8e9b3af
Compare
Contributor
There was a problem hiding this comment.
Actionable comments posted: 2
🤖 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 17-21: Update the description for the deepmd-train skill so it no
longer implies only training from scratch: change the opening sentence to say
the skill handles training and fine‑tuning workflows (including from‑scratch and
pretrained/transfer-style DPA3 paths), and keep the rest of the paragraph about
progressive disclosure and model-specific configs under
skills/deepmd-train/models/ intact.
In `@skills/deepmd-train/models/se-e2-a.md`:
- Around line 12-16: The example for computing neighbor selection `sel`
hard-codes element types with `-t O H`; update the `dp --pt neighbor-stat`
example so it uses the dataset's actual type_map instead of fixed "O H" (e.g.
show `-t <type_map>` or omit `-t` and explain how to extract the dataset's
type_map) and mention `type_map` by name; change the line in se-e2-a.md that
currently shows `dp --pt neighbor-stat -s /path/to/data -r 6.0 -t O H` to a
model-agnostic form like `dp --pt neighbor-stat -s /path/to/data -r 6.0 -t
<type_map>` (or `dp --pt neighbor-stat -s /path/to/data -r 6.0` with a note how
to get the dataset's type_map), and ensure the surrounding text references `sel`
and `type_map` so users know to substitute their dataset's types.
🪄 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: 673d9a2c-056d-49bb-8dd2-e0db695947ab
📒 Files selected for processing (6)
doc/agent-skills.mdskills/deepmd-train-dpa3/SKILL.mdskills/deepmd-train-se-e2-a/SKILL.mdskills/deepmd-train/SKILL.mdskills/deepmd-train/models/dpa3.mdskills/deepmd-train/models/se-e2-a.md
💤 Files with no reviewable changes (2)
- skills/deepmd-train-se-e2-a/SKILL.md
- skills/deepmd-train-dpa3/SKILL.md
✅ Files skipped from review due to trivial changes (1)
- skills/deepmd-train/models/dpa3.md
njzjz
force-pushed
the
skill-progressive-train
branch
from
8e9b3af to
4ab0212
Compare
Contributor
There was a problem hiding this comment.
🧹 Nitpick comments (2)
skills/deepmd-train/SKILL.md (2)
119-119: 💤 Low valueClarify what "clearly marked" means for placeholder data paths.
Line 119 expects data paths to be "clearly marked as placeholders," but the checklist doesn't specify the format or convention. Consider adding an example (e.g.,
/path/to/train_systemor<PLACEHOLDER: training data>) to guide agents on acceptable patterns.🤖 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/SKILL.md` at line 119, Update the checklist item "Training/validation data paths exist or are clearly marked as placeholders" to define the placeholder convention and give examples; specifically, modify the SKILL.md checklist entry to state what "clearly marked" means (for example: use a distinct pattern like /path/to/train_system or <PLACEHOLDER: training data>), allow either absolute paths or the placeholder syntax, and include one or two concrete examples so agents know acceptable formats (e.g., "/path/to/train_system" and "<PLACEHOLDER: training data>").
26-28: 💤 Low valueReduce repetition of "If" at sentence starts.
Lines 26–28 each begin with "If," which LanguageTool flagged as repetitive. Consider reordering or merging to improve readability while preserving the logical structure.
Example rephrase:
- Classify based on the user's input: if they already named a model, read only that reference; if they ask for a recommendation, collect decision inputs and choose before reading; if model-specific parameters aren't needed yet, stay in the top-level workflow.
🤖 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/SKILL.md` around lines 26 - 28, Combine the three short sentences that each start with "If" into a single, clearer sentence to reduce repetition and preserve logic; for example, in SKILL.md's workflow description replace the three lines beginning "If the user..." / "If the user asks..." / "If model-specific..." with a single sentence such as: "Classify based on the user's input: if they already named a model, read only that reference; if they ask for a recommendation, collect the decision inputs below, choose a model, then read only the selected reference; if model-specific parameters aren't needed yet, stay in this top-level workflow." Ensure the revised sentence retains the original ordering and meaning of the three cases.Source: Linters/SAST tools
🤖 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 `@skills/deepmd-train/SKILL.md`:
- Line 119: Update the checklist item "Training/validation data paths exist or
are clearly marked as placeholders" to define the placeholder convention and
give examples; specifically, modify the SKILL.md checklist entry to state what
"clearly marked" means (for example: use a distinct pattern like
/path/to/train_system or <PLACEHOLDER: training data>), allow either absolute
paths or the placeholder syntax, and include one or two concrete examples so
agents know acceptable formats (e.g., "/path/to/train_system" and "<PLACEHOLDER:
training data>").
- Around line 26-28: Combine the three short sentences that each start with "If"
into a single, clearer sentence to reduce repetition and preserve logic; for
example, in SKILL.md's workflow description replace the three lines beginning
"If the user..." / "If the user asks..." / "If model-specific..." with a single
sentence such as: "Classify based on the user's input: if they already named a
model, read only that reference; if they ask for a recommendation, collect the
decision inputs below, choose a model, then read only the selected reference; if
model-specific parameters aren't needed yet, stay in this top-level workflow."
Ensure the revised sentence retains the original ordering and meaning of the
three cases.
ℹ️ Review info
⚙️ Run configuration
Configuration used: Repository UI
Review profile: CHILL
Plan: Pro
Run ID: 7c7ee886-eeac-443a-bf0d-f84a2e453f1a
📒 Files selected for processing (6)
doc/agent-skills.mdskills/deepmd-train-dpa3/SKILL.mdskills/deepmd-train-se-e2-a/SKILL.mdskills/deepmd-train/SKILL.mdskills/deepmd-train/models/dpa3.mdskills/deepmd-train/models/se-e2-a.md
💤 Files with no reviewable changes (2)
- skills/deepmd-train-se-e2-a/SKILL.md
- skills/deepmd-train-dpa3/SKILL.md
✅ Files skipped from review due to trivial changes (1)
- doc/agent-skills.md
🚧 Files skipped from review as they are similar to previous changes (2)
- skills/deepmd-train/models/se-e2-a.md
- skills/deepmd-train/models/dpa3.md
Move model-specific DeePMD-kit training recipes under a single deepmd-train skill so agents first choose a model and only then read the selected configuration reference. Document the progressive-disclosure pattern for future skill additions. Authored by OpenClaw (model: custom-chat-jinzhezeng-group/gpt-5.5)
njzjz
force-pushed
the
skill-progressive-train
branch
from
4ab0212 to
a9dfc01
Compare
Codecov Report✅ All modified and coverable lines are covered by tests. Additional details and impacted files@@ Coverage Diff @@
## master #5530 +/- ##
==========================================
- Coverage 82.19% 82.19% -0.01%
==========================================
Files 891 891
Lines 101599 101599
Branches 4242 4240 -2
==========================================
- Hits 83507 83506 -1
Misses 16789 16789
- Partials 1303 1304 +1 ☔ View full report in Codecov by Harness. 🚀 New features to boost your workflow:
|
iProzd
approved these changes
Jun 15, 2026
njzjz
approved these changes
Jun 16, 2026
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
3 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
deepmd-trainskill that first selects the model family, then reads only the chosen model reference underskills/deepmd-train/models/.doc/agent-skills.mdpage to list the consolidated training skill and adjust the verification example.Notes
pre-commit run --files doc/agent-skills.md skills/deepmd-train/SKILL.md skills/deepmd-train/models/dpa3.md skills/deepmd-train/models/se-e2-a.md; all selected hooks passed.se_e2_aspelling is lowercase.sphinxis not installed in this checkout environment.Authored by OpenClaw (model: custom-chat-jinzhezeng-group/gpt-5.5)
Summary by CodeRabbit