Performs safe code refactoring with comprehensive safety measures.

Per-agent evolve/skip decision maker — triggers factory pipeline for agents needing improvement.

>

Improves typography by fixing font choices, hierarchy, sizing, weight, and readability so text feels intentional. Use when the user mentions fonts, type, readability, text hierarchy, sizing looks off, or wants more polished, intentional typography.

Designs and improves onboarding flows, empty states, and first-run experiences to help users reach value quickly. Use when the user mentions onboarding, first-time users, empty states, activation, getting started, or new user flows.

Work through every step in order. Record "no matches" and continue when a step finds nothing.

This document collects ideas and instructions for implementing future improvements. Follow these when adding features or refactoring the code.

**Scope:** `src-tauri/src/db/` - SQLite repository pattern

- **ALWAYS bump the version in `package.json`** after every change merged to main. Don't wait for the user to ask. Use semver: patch for fixes/refactors, minor for features, major for breaking changes

Launch N parallel research agents that explore different improvement approaches, cross-pollinate ideas, and converge on a unified plan

- Use explicit primary keys and foreign keys

Use **conventional commits** for all changes:

Use **conventional commits** for all changes:

Use conventional commits:

When the user wants concrete recommendations on how to improve their social media performance. Also use when the user mentions 'what should I do next,' 'how do I improve,' 'optimize my social media,' 'recommendations,' 'suggestions,' 'next steps,' 'what's my biggest opportunity,' or 'help me grow.' Synthesizes insights from performance, audience, and pattern analysis into prioritized actions. For raw analytics, see performance-analyzer-sms. For growth tracking, see audience-growth-tracker-sms. For pattern detection, see content-pattern-analyzer-sms.

Restructure the "💥 Todo: Opez" note to match the standard todo format.

Improve unclear UX copy, error messages, microcopy, labels, and instructions to make interfaces easier to understand. Use when the user mentions confusing text, unclear labels, bad error messages, hard-to-follow instructions, or wanting better UX writing.

Audit and refactor CSS to comply with Game Loopers design system and BEM methodology

Check SI notebooks in notebooks/manuscripts/ for deprecated API usage, hardcoded paths, stale parameter names, and compatibility with the current TEXAS package. Use after refactors or before paper submission.

Autonomous end-to-end execution skill. Use when the user gives a broad directive (improve, audit, fix, ship, polish) that should run from audit through deploy without check-ins. This skill codifies the loop-until-live pattern and the stop conditions.

Ensures new Rails views, layouts, and page-level UI in rails-pokedex are designed in Google Stitch and retrieved via MCP before coding. Use when creating ERB templates, layouts, partials, styling overhauls, or any new user-facing page; when refactoring views; or when the user mentions Stitch, UI de página, telas, or frontend Rails.

Research and improve Claude skills with current best practices. Triggers on requests to improve skills, update skills, research best practices for skills, enhance skill quality, or modernize existing skills.

>

Guide for creating React components following project conventions — memo pattern, TypeScript props, Tailwind styling, accessibility. Use when creating new components, refactoring existing ones, or when the user asks about component patterns.

Руководство по созданию и управлению миграциями баз данных в SS14 (PostgreSQL и SQLite)

Explain how a file, folder, or feature works

Map the repository before major refactors. Explain architecture, ownership, data flow, control flow, and blast radius in plain English.

`web/`의 Prisma 7 설정과 generated client import 규칙을 맞추기 위한 스킬이다. 스키마 변경, migration, import 패턴 정리, Prisma 타입 사용 시 적용한다.

Fast semantic code navigation via LSP. Load FIRST before ANY code task - even 'simple' ones. Trigger scenarios: (1) fixing lint/type/pyright/mypy warnings or errors, (2) fixing reportAny/reportUnknownType/Any type errors, (3) adding type annotations, (4) refactoring or modifying code, (5) finding where a function/class/symbol is defined, (6) finding where a symbol is used/referenced/imported, (7) understanding what a function calls or what calls it, (8) exploring unfamiliar code or understanding architecture, (9) renaming symbols across codebase, (10) finding interface/protocol implementations, (11) ANY task where you'd use ripgrep to find code or read-file to view a function. Use `leta show SYMBOL` instead of read-file, `leta refs SYMBOL` instead of ripgrep for usages, `leta grep PATTERN` instead of ripgrep for definitions, `leta files` instead of list-directory.

Self-improving agent architecture using ChromaDB for continuous learning, self-evaluation, and improvement storage. Agents maintain separate memory collections for learned patterns, performance metrics, and self-assessments without modifying their static .md configuration.

Manages deprecation and migration. Use when removing old systems, APIs, or features. Use when migrating users from one implementation to another. Use when deciding whether to maintain or sunset existing code.

Generate user-facing release notes from tickets, PRDs, or changelogs. Creates clear, engaging summaries organized by category (new features, improvements, fixes). Use when writing release notes, creating changelogs, announcing product updates, or summarizing what shipped.

Enforce British English using Oxford spelling conventions, permit outwith and caveat where appropriate, and use the Oxford comma only when it improves clarity.

Provides an incremental adoption strategy to migrate native iOS or Android apps to React Native or Expo using @callstack/react-native-brownfield for initial setup. Use when planning migration steps, packaging XCFramework/AAR artifacts, and integrating them into host apps.

Generate an Alembic migration with proper naming and structure

App architecture and implementation planning principles — screen structure, data models, phase breakdown, UI pattern decisions, and command recommendations. Auto-load when planning app features or architecture.

In-app notifications (SQLite rows, no realtime): kinds, emit sites, inbox/API. Use when adding triggers, kinds, or notification UI/API work.

Compile rolling daily briefs into a convergence-led narrative with content outputs and explicit accept/defer/reject decisions for recursive learning.

Improves UI/UX in this Next.js + Tailwind app by running a structured design pass (IA, flows, accessibility, responsive, components). Use when the user asks for UI polish, UX improvements, redesign, landing page tweaks, component library work, Tailwind styling, accessibility fixes, or visual consistency.

springdoc/OpenAPI sync, runbooks, changelogs, migration guides, and architecture diagram updates.

Generates git commit messages following this repository's conventions. Analyzes staged changes and outputs properly formatted commit messages. Use when user asks to write, generate, or format a commit message.

Plan first and request user confirmation before implementation. Use when the user wants explicit validation before code changes, risky refactors, migrations, or any multi-step change where approval is required before building.

Prisma migrations, file storage, SSRF protection, and secrets handling rules.

Analyze references, citations, and cluster structure to improve reading order and synthesis.

Improve dense PM Workspace tables and forms such as registers, tracker, tasks, and timesheets. Use for inline editing, mobile fallbacks, direct actions, and clearer dense-data workflows.

'Push the LLM to reconsider, refine, and improve recent output through iterative elicitation methods. Use when called from a Compass workflow checkpoint.'

Create clean, conventional git commits

Use when writing prompts that generate better prompts. A meta-prompt that improves any AI conversation.

从自然语言数据模型描述生成 Supabase 全栈项目代码（migration、RLS、API、hooks）

从数据库 schema 定义自动生成文档和 Mermaid ERD 图