探索开发、内容、数据与办公效率等 AI 技能。先比较用途和仓库热度，再查看技能说明与来源。
Use when receiving code review feedback, before implementing suggestions, especially if feedback seems unclear or technically questionable - requires technical rigor and verification, not performative agreement...
search the web for information when user explicitly requests web searches, by identifying the nature of the request and delegating to appropriate subagents (@code-researcher for documentation/code examples, @we...
Determine package manager, workspace configuration, and available commands in TypeScript/JavaScript projects. Use when agent needs to understand how to run commands in a TS/JS project, including (1) Detecting...
Guide for diagnosing lsp errors. This skill should be used when users wants to debug a lsp/diagnostic errors, warnings, or exceptions and get clear explanations and potential solutions. "Why am i getting thi...
Generate explanatory lessons after completing speclet stories to develop mechanical coding fluency
Convert a draft document into an executable spec.json with trackable stories
Run parallel reviews on draft/spec/ticket inputs and emit Council artifacts
Run parallel multi-model reviews on .speclet/draft.md and emit Council artifacts
Execute one autonomous iteration - implement the next story from spec.json
Generate a collaborative draft document with clarifying questions for a new feature
Merge accepted council feedback into draft.md and keep review artifacts separate
Generate a collaborative draft document with clarifying questions for a new feature
Deprecated. Use /ralph-loop instead.
Convert a draft document into an executable spec.json with trackable stories
Convert a large draft into individual tickets for isolated implementation sessions
Run the complete PDF processing pipeline to convert the manual PDF into Next.js application data. This command executes all steps automatically: split, render, extract, translate, build, and manifest.
Capture all manual pages and verify translations match page images using AI-powered checking. Identifies extraction failures and can regenerate extracted text from images.
Capture screenshots of all manual pages at high resolution for translation verification. Use when verifying that translations match page images, checking layout correctness, or inspecting for content mismatches...
A permanent record of identified bugs, architectural pitfalls, and logic errors. MUST be referenced before major refactors.
Defines the standard for peer-to-peer communication between Antigravity (Frontend Agent) and other specialized agents (e.g., Backend IDE Agent).
Automated delivery pipeline: Static Check -> Logic Test -> Smoke Test -> Deploy.
Official Design Language for "Just Mowing" - Baba Is You Meets Wasteland.
Perform a comprehensive, cognitive static analysis of a specific feature or logic flow to identify logical flaws, missing connections, or design inconsistencies.
Enforces the Single Source of Truth (SSOT) and directory classification rules for the "Grass Cutting" game project.
默认按已收录数据中的 GitHub 仓库 Stars 排序。Stars 属于来源仓库，同一仓库的多个技能可能共享相同数值；它反映仓库关注度，不代表单个技能的安装量或质量评分。