All categories
Code Review & Quality
Code review, refactoring, debugging, maintainability, and engineering quality
- 15,957
- 665
Security grades appear on each card once the skill has been scanned. Newly imported skills may briefly show without a grade until the backfill job runs.
Open in full browserBrowse code review & quality skills
Showing 11,137–11,160 of 15,957 skills
- TodosGuidelines for managing the Claude Code todo list during sessions.Votes: 0GitHub stars: 52
- TaskQuick-create a task linked to a project, optionally with subtasks.Votes: 0GitHub stars: 52
- Tag Management**Purpose:** Analyze, refactor, and maintain tags across the vault according to the official taxonomy. **Taxonomy Reference:** [[.claude/rules/tag-taxonomy.md]] ---Votes: 0GitHub stars: 52
- Skill CreatorCreate new skills with guided prompts following best practicesVotes: 0GitHub stars: 52
- De Ai IfyRewrite AI-generated text to sound natural and human-writtenVotes: 0GitHub stars: 52
- C4 DiagramGenerate C4 architecture diagrams from system notesVotes: 0GitHub stars: 52
- Broken LinksScan the vault for broken [[wiki-links]] and suggest fixes using parallel Sonnet sub-agents.Votes: 0GitHub stars: 52
- Auto TagBatch auto-tag untagged notes using content analysisVotes: 0GitHub stars: 52
- Writing SkillsUse when checking, testing, or reviewing existing skills for trigger accuracy, environment compatibility, or deployment readiness. Not for learning how to write skills from scratch.Votes: 0GitHub stars: 2
- Using SuperpowersUse this skill at the beginning of EVERY user request to check for specialized skills before acting. Invoke immediately when the user mentions coding, debugging, writing, analyzing, reviewing, testing, designing, or any specific task—before asking clarifying questions or taking any action.Votes: 0GitHub stars: 2
- BrainstormingUse this skill when the user wants to discuss, plan, or design something BEFORE implementation. Triggers for: exploring ideas, gathering requirements, creating design specs, comparing architectural approaches, or reviewing design decisions. Does NOT trigger for direct implementation requests like 'write code', 'generate a component', 'create a document', 'fix a bug', or 'deploy this' - those skip straight to implementation. The key signal is the user wants to THINK THROUGH and DESIGN somethin...Votes: 0GitHub stars: 2
- Aily Self ImprovementCaptures learnings, errors, and corrections to enable continuous improvement. Use when: (1) An operation or tool fails unexpectedly, (2) User corrects the agent, (3) User requests a capability that doesn't exist, (4) A better approach or knowledge gap is discovered. Also review learnings before major tasks.Votes: 0GitHub stars: 2
- Aily Doc Project Kickoff Report项目启动报告：统一目标、范围、里程碑、分工与风险，形成可落地的启动材料。Votes: 0GitHub stars: 2
- Review ChangesReview all uncommitted changes in the current repo and produce a structured code review report.Votes: 0GitHub stars: 2
- Model Architecture"Inspect and modify YOLOv3 model YAMLs, anchors, Detect heads,Votes: 0GitHub stars: 247
- Yolov3"Operate Ultralytics YOLOv3 detection workflows: inference,Votes: 0GitHub stars: 247
- Evaluation"Enable built-in perplexity evaluation with `model.evaluate` andVotes: 0GitHub stars: 247
- Model Modules"Add and debug VLM-R1 model modules, including Qwen2VL, InternVL,Votes: 0GitHub stars: 247
- Data And Rewards"Prepare and validate verl post-training parquet data, reward-modelVotes: 0GitHub stars: 247
- Scalar Quantizers"Operate FSQ, FSP, ResidualFSQ, and GroupedResidualFSQ scalarVotes: 0GitHub stars: 247
- Salient Object Inference"Run U-2-Net salient object and human segmentation inference fromVotes: 0GitHub stars: 247
- Data And Training"Route U-2-Net dataset validation, transform inspection, and safeVotes: 0GitHub stars: 247
- U 2 Net"Route U-2-Net salient object detection, human segmentation,Votes: 0GitHub stars: 247
- Training"Training, configuration, data, checkpointing, PEFT, and sweepVotes: 0GitHub stars: 247