veomni-review
Use this skill before committing ANY code change — this is a mandatory gate in the commit flow. Also trigger proactively when: you've made changes across multiple files and want to check consistency, you're unsure if a fix is safe, a change touches shared infrastructure (BaseTrainer, distributed, model loading, data pipeline), or a change is larger than a few lines. The review launches a subagent that checks implementation quality, multi-file consistency, and known constraint violations, then rates the change as safe/needs-attention/risky.
Best use case
veomni-review is best used when you need a repeatable AI agent workflow instead of a one-off prompt.
Use this skill before committing ANY code change — this is a mandatory gate in the commit flow. Also trigger proactively when: you've made changes across multiple files and want to check consistency, you're unsure if a fix is safe, a change touches shared infrastructure (BaseTrainer, distributed, model loading, data pipeline), or a change is larger than a few lines. The review launches a subagent that checks implementation quality, multi-file consistency, and known constraint violations, then rates the change as safe/needs-attention/risky.
Teams using veomni-review should expect a more consistent output, faster repeated execution, less prompt rewriting.
When to use this skill
- You want a reusable workflow that can be run more than once with consistent structure.
When not to use this skill
- You only need a quick one-off answer and do not need a reusable workflow.
- You cannot install or maintain the underlying files, dependencies, or repository context.
Installation
Claude Code / Cursor / Codex
Manual Installation
- Download SKILL.md from GitHub
- Place it in
.claude/skills/veomni-review/SKILL.mdinside your project - Restart your AI agent — it will auto-discover the skill
How veomni-review Compares
| Feature / Agent | veomni-review | Standard Approach |
|---|---|---|
| Platform Support | Not specified | Limited / Varies |
| Context Awareness | High | Baseline |
| Installation Complexity | Unknown | N/A |
Frequently Asked Questions
What does this skill do?
Use this skill before committing ANY code change — this is a mandatory gate in the commit flow. Also trigger proactively when: you've made changes across multiple files and want to check consistency, you're unsure if a fix is safe, a change touches shared infrastructure (BaseTrainer, distributed, model loading, data pipeline), or a change is larger than a few lines. The review launches a subagent that checks implementation quality, multi-file consistency, and known constraint violations, then rates the change as safe/needs-attention/risky.
Where can I find the source code?
You can find the source code on GitHub using the link provided at the top of the page.
Related Guides
AI Agents for Coding
Browse AI agent skills for coding, debugging, testing, refactoring, code review, and developer workflows across Claude, Cursor, and Codex.
Best AI Skills for Claude
Explore the best AI skills for Claude and Claude Code across coding, research, workflow automation, documentation, and agent operations.
Cursor vs Codex for AI Workflows
Compare Cursor and Codex for AI coding workflows, repository assistance, debugging, refactoring, and reusable developer skills.
SKILL.md Source
## Steps
1. Run `git diff` (staged + unstaged) to capture the full diff.
2. Read `.agents/knowledge/constraints.md` for known constraints.
3. **Launch a review subagent** (see below). The subagent receives only the diff + constraints — NOT your reasoning — to avoid confirmation bias.
4. Act on the verdict.
| Verdict | Action |
|---------|--------|
| **safe** | Proceed to commit |
| **needs-attention** | Address listed issues, then commit |
| **risky** | Output the report, do NOT commit, wait for user |
5. Run `make quality` before the final commit.
## Subagent Launch
Use the Task tool with this prompt:
```
You are a code reviewer for VeOmni, a distributed multi-modality training framework. Your job is to find problems in the following diff. You are NOT validating the author's intent — you are looking for bugs, risks, and constraint violations.
## Diff
<paste full git diff here>
## Known Constraints
<paste constraints.md content here>
## Review Checklist
For each changed file, check:
### Implementation Quality
- Hidden risks or edge cases not handled?
- Simpler alternative that achieves the same result?
- Boundary conditions (tensor shapes, distributed rank handling, gradient accumulation steps)?
- Does the fix depend on downstream code to "clean up"?
### Multi-file Consistency
- If a Trainer method changed, do all subclasses need matching changes?
- If model loading changed, are configs and parallel plans updated?
- If data collator changed, do all modalities still work?
- If distributed code changed, are both FSDP and FSDP2 paths handled?
### Constraint Violations
- Does this violate any entry in the known-constraints list?
- Does this repeat a pattern that previously caused bugs?
### VeOmni-Specific Checks
- PR title format: `[{modules}] {type}: {description}`?
- All comments and docstrings in English?
- No auto-generated files (`veomni/models/transformers/*/generated/`) edited directly?
- Ruff-compliant (`make quality` passes)?
## Output
### Verdict: safe / needs-attention / risky
### Findings (for needs-attention or risky)
For each issue:
- **File**: path:line
- **Concern**: what could go wrong
- **Suggestion**: what to do instead
```
## After Commit
- Run `make quality` to confirm ruff compliance.
- Verify PR title follows `[{modules}] {type}: {description}` format.Related Skills
veomni-uv-update
Use this skill when updating dependencies managed by uv: bumping a package version, upgrading the uv tool itself, updating torch/CUDA stack, switching transformers version, or regenerating the lockfile. Trigger: 'update dependency', 'bump version', 'upgrade uv', 'update torch', 'update lockfile', 'uv sync fails'.
veomni-new-op
Use this skill when adding a new optimized kernel or operator to veomni/ops/. Covers the full lifecycle: understanding VeOmni's ops architecture (monkey-patch + global function pointer pattern), implementing the kernel, registering it, adding tests, and documenting it. Trigger: 'add op', 'new kernel', 'add attention variant', 'new fused op', 'add triton kernel', 'optimize operator'.
veomni-new-model
Use this skill when adding support for a new model to VeOmni. Covers the full lifecycle: analyzing the HuggingFace model, creating model patches, defining parallel plans, writing configs, integrating with the trainer, and testing. Trigger: 'add model', 'support new model', 'integrate <model_name>', 'new model support'.
veomni-develop
VeOmni-specific checklist for feature development and refactoring. Covers impact analysis across modalities, trainer hierarchy, data pipeline, and distributed code. Use before implementing any non-trivial change. For model-specific or ops-specific work, use veomni-new-model or veomni-new-op instead. Trigger: 'add feature', 'implement', 'refactor', 'reorganize', 'new capability'.
veomni-debug
Use this skill for ANY bug, error, crash, wrong output, loss divergence, gradient explosion, test failure, CUDA error, distributed training hang, checkpoint load failure, or unexpected behavior. Covers both quick fixes (clear root cause) and complex debugging (unclear cause). Trigger: 'fix bug', 'fix error', 'broken', 'crash', 'doesn't work', 'fails with', 'loss NaN', 'training hangs', 'FSDP error', 'OOM'.
create-pr
Create a pull request for the current branch. Handles uncommitted changes, generates a PR title matching the `[{modules}] {type}: {description}` format enforced by CI, and fills in the PR description template. Trigger: 'create pr', 'open pr', 'submit pr', 'make pr'.
flutter-dart-code-review
库无关的Flutter/Dart代码审查清单，涵盖Widget最佳实践、状态管理模式（BLoC、Riverpod、Provider、GetX、MobX、Signals）、Dart惯用法、性能、可访问性、安全性和整洁架构。
security-review
Use this skill when adding authentication, handling user input, working with secrets, creating API endpoints, or implementing payment/sensitive features. Provides comprehensive security checklist and patterns.
addressing-pr-review-comments
Address all valid review comments on a PR for the current branch in the streamlit/streamlit repo. Covers both inline review comments and general PR (issue) comments. Use when a PR has reviewer feedback to address, including code changes, style fixes, and documentation updates.
lightning-architecture-review
Review Bitcoin Lightning Network protocol designs, compare channel factory approaches, and analyze Layer 2 scaling tradeoffs. Covers trust models, on-chain footprint, consensus requirements, HTLC/PTLC compatibility, liveness, and watchtower support.
gha-security-review
Find exploitable vulnerabilities in GitHub Actions workflows. Every finding MUST include a concrete exploitation scenario — if you can't build the attack, don't report it.
gh-review-requests
Fetch unread GitHub notifications for open PRs where review is requested from a specified team or opened by a team member. Use when asked to "find PRs I need to review", "show my review requests", "what needs my review", "fetch GitHub review requests", or "check team review queue".