Davie521

@davie521 source repo

49 skills
0 followers
10 hours ago last updated

▌
Cleanup · davie521
End-of-session cleanup: shut down every process the session started, find code-vs-doc mismatches and update the DOCS to match the code (code is the source of truth; never rewrite code just to match a doc), drive leftover tasks to zero (finish or explicitly report), and persist lessons learned to memory. Each phase must end with an explicit done / not-needed verdict — no silent skipping. Use when the user says 'cleanup', '收尾', '清理现场', '打扫', wants to wrap up a work session, or before handing off / ending a long task.

0 installs
▌
Santa Method · davie521
Multi-agent adversarial verification with convergence loop. Two independent review agents must both pass before output ships.

0 installs
▌
Tdd Workflow · davie521
Use this skill when writing new features, fixing bugs, or refactoring code. Enforces test-driven development with 80%+ coverage including unit, integration, and E2E tests.

0 installs
▌
Rules Distill · davie521 bundle
Scan skills to extract cross-cutting principles and distill them into rules — append, revise, or create new rule files

0 installs
▌
Vibe Writing · davie521 bundle
AI-assisted writing workflow with four phases: Learn (Socratic dialogue → knowledge cards), Structure (article framework), Write (expand and polish), Finalize (assemble full article). Use when user says '写作', '写文章', '帮我写', 'vibe', or wants to co-author any written content.

0 installs
▌
Diagnose · davie521 bundle
Disciplined diagnosis loop for hard bugs and performance regressions. Reproduce → minimise → hypothesise → instrument → fix → regression-test. Use when user says "diagnose this" / "debug this", reports a bug, says something is broken/throwing/failing, or describes a performance regression.

0 installs
▌
Security Scan · davie521
Scan your Claude Code configuration (.claude/ directory) for security vulnerabilities, misconfigurations, and injection risks using AgentShield. Checks CLAUDE.md, settings.json, MCP servers, hooks, and agent definitions.

0 installs
▌
Video Editing · davie521
AI-assisted video editing workflows for cutting, structuring, and augmenting real footage. Covers the full pipeline from raw capture through FFmpeg, Remotion, ElevenLabs, fal.ai, and final polish in Descript or CapCut. Use when the user wants to edit video, cut footage, create vlogs, or build video content.

0 installs
▌
Design Challenge · davie521
Challenge an implementation approach or design choice via Codex (read-only `codex exec` through the codex-run wrapper). Read-only, asks "is this the right path?" — not "are there bugs?". Trigger when the user says "质疑这个方案" / "这个设计靠不靠谱" / "challenge this design" / "design review" / "second-guess this approach" / "考虑过 X 吗" / "is this the right approach" / "stress-test the design" / "punch holes in this". Returns assumptions / failure modes / alternatives — NOT a bug list, NOT a severity matrix. Division of labor with grill-with-docs — user wants to be interactively grilled question-by-question → dev-workflow:grill-with-docs; user wants a fully-formed proposal attacked by an external model in one pass → this skill.

0 installs
▌
Deep Plan · davie521
Restate requirements, assess risks, create a step-by-step implementation plan, and WAIT for explicit confirmation before touching any code. On approval the FIRST action MUST be creating an isolated git worktree (never edit the main worktree); implement inside it; after implementation hand off to /cpr (PR pipeline) or code review. Use when starting a feature, architectural change, complex refactor, multi-file change, or when requirements are unclear.

0 installs
▌
Prototype · davie521 bundle
Build a throwaway prototype to flush out a design before committing to it. Routes between two branches — a runnable terminal app for state/business-logic questions, or several radically different UI variations toggleable from one route. Use ONLY for two scenarios — (1) the user needs to compare multiple structurally different UI approaches side by side before picking one, or (2) the user needs to manually step through a state machine / data model that is hard to reason about on paper. For building a real page or polished UI, use ui-ux-pro-max / hallmark instead — this skill produces throwaway code.

0 installs
▌
Skill Stocktake · davie521 bundle
Use when auditing Claude skills and commands for quality. Supports Quick Scan (changed skills only) and Full Stocktake modes with sequential subagent batch evaluation.

0 installs
▌
Swiftui Patterns · davie521
SwiftUI architecture patterns, state management with @Observable, view composition, navigation, actor-based persistence, protocol-based DI for testing, performance optimization, and modern iOS/macOS UI best practices.

0 installs
▌
Data Scraper Agent · davie521
Build a fully automated AI-powered data collection agent for any public source — job boards, prices, news, GitHub, sports, anything. Scrapes on a schedule, enriches data with a free LLM (Gemini Flash), stores results in Notion/Sheets/Supabase, and learns from user feedback. Runs 100% free on GitHub Actions. Use when the user wants to monitor, collect, or track any public data automatically.

0 installs
▌
Search Routing · davie521
Single decision skill for all web search needs. exa is the default search backend (including `site:` and Chinese queries); firecrawl is for scraping pages exa cannot fetch, and firecrawl_search is only a fallback. Asks user before invoking expensive deep modes. Use this BEFORE any web search call when the user has multiple search MCPs available.

0 installs
▌
Deployment Patterns · davie521
Deployment strategies (rolling, blue-green, canary), CI/CD pipeline patterns, health checks and Kubernetes probes, rollback playbooks, and production readiness checklists for web applications. Use when planning a release, setting up CI/CD, or preparing rollback. Dockerfile and Compose specifics live in docker-patterns.

0 installs
▌
MCP Server Patterns · davie521
Build MCP servers with Node/TypeScript SDK — tools, resources, prompts, Zod validation, stdio vs Streamable HTTP. Use when the user wants to write, extend, or debug an MCP server — implementing tools/resources/prompts, choosing a transport, upgrading the SDK, or fixing registration/transport issues. Use Context7 or official MCP docs for latest API.

0 installs
▌
Database Migrations · davie521
Database migration best practices for schema changes, data migrations, rollbacks, and zero-downtime deployments in PostgreSQL with Prisma, Drizzle, Kysely, and Alembic — plus a PostgreSQL quick reference for query optimization, indexing, schema design, and Row Level Security. Use when writing SQL queries or migrations, designing Postgres schemas or indexes, troubleshooting slow queries, implementing RLS, or planning zero-downtime schema changes.

0 installs
▌
Doc Coauthoring · davie521
Guide users through a structured workflow for co-authoring documentation. Use when user wants to write documentation, proposals, technical specs, decision docs, or similar structured content. This workflow helps users efficiently transfer context, refine content through iteration, and verify the doc works for readers. Trigger when user mentions writing docs, creating proposals, drafting specs, or similar documentation tasks.

0 installs
▌
Click Path Audit · davie521
Trace every user-facing button/touchpoint through its full state change sequence to find bugs where functions individually work but cancel each other out, produce wrong final state, or leave the UI in an inconsistent state. Use when: systematic debugging found no bugs but users report broken buttons, or after any major refactor touching shared state stores.

0 installs
▌
Liquid Glass Design · davie521
Apple native iOS 26 (SwiftUI/UIKit/WidgetKit) — NOT web CSS glassmorphism/backdrop-filter. iOS 26 Liquid Glass design system — dynamic glass material with blur, reflection, and interactive morphing via glassEffect/UIGlassEffect. Use when building or migrating native Apple apps to the iOS 26 glass design language; for glassmorphism on web pages use ui-ux-pro-max instead.

0 installs
▌
Security Checklist · davie521 bundle
Security review checklist reference — not the built-in /security-review command. 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.

0 installs
▌
Investor Materials · davie521
Create and update pitch decks, one-pagers, investor memos, accelerator applications, financial models, fundraising materials, and investor outreach (cold emails, warm intro blurbs, follow-ups). Use when the user needs investor-facing documents, projections, use-of-funds tables, milestone plans, outreach or follow-up emails to angels/VCs/accelerators, or materials that must stay internally consistent across multiple fundraising assets.

0 installs
▌
Codebase To Course · davie521 bundle
Turn any codebase into a beautiful, interactive single-page HTML course that teaches how the code works to non-technical people. Use this skill whenever someone wants to create an interactive course, tutorial, or educational walkthrough from a codebase or project. Also trigger when users mention 'turn this into a course,' 'explain this codebase interactively,' 'teach this code,' 'interactive tutorial from code,' 'codebase walkthrough,' 'learn from this codebase,' or 'make a course from this project.' This skill produces a stunning, self-contained HTML file with scroll-based navigation, animated visualizations, embedded quizzes, and code-with-plain-English side-by-side translations.

0 installs
▌
Visa Doc Translate · davie521
Produce an UNCERTIFIED draft English translation of visa application document images as a bilingual PDF (original + translation). Output is a draft for a human translator to review and certify — it is not a certified translation and must not be filed as one.

0 installs
▌
Swift Concurrency 6 2 · davie521
Swift 6.2 Approachable Concurrency — single-threaded by default, @concurrent for explicit background offloading, isolated conformances for main actor types. Use when migrating a project to Swift 6.2 concurrency, adopting @concurrent or MainActor-default inference, or resolving data-race safety compiler errors ("sending X risks causing data races").

0 installs
▌
Grill With Docs · davie521 bundle
Grilling session that challenges your plan against the existing domain model, sharpens terminology, and updates documentation (CONTEXT.md, ADRs) inline as decisions crystallise. Use when user wants to stress-test a plan, get grilled on their design, mentions "grill me", or wants to challenge a plan against their project's language and documented decisions.

0 installs
▌
LLM Cost Discipline · davie521
Cost discipline for LLM pipelines — parse structured text with regex first and escalate only flagged items to an LLM, route the cheapest sufficient model tier by task size, track budget, retry narrowly, cache prompts. Use when building LLM pipelines, controlling API costs, or choosing between regex and LLM for parsing structured text (构建 LLM 管道、控制 API 成本、解析结构化文本选方案时).

0 installs
▌
Plankton Code Quality · davie521
Write-time code quality enforcement using Plankton — auto-formatting, linting, and Claude-powered fixes on every file edit via hooks.

0 installs
▌
Content Hash Cache Pattern · davie521
Cache expensive file processing results using a composite key of SHA-256 content hash plus processor version and config — path-independent, auto-invalidating, with service layer separation. Use when caching costly file-processing outputs (PDF parsing, OCR, text extraction, image analysis) that recur across runs and the cache must survive file moves/renames while auto-invalidating on content change.

0 installs
▌
Foundation Models On Device · davie521
Apple FoundationModels framework for on-device LLM — text generation, guided generation with @Generable, tool calling, and snapshot streaming in iOS 26+.

0 installs
▌
Improve Codebase Architecture · davie521 bundle
Find deepening opportunities in a codebase, informed by the domain language in CONTEXT.md and the decisions in docs/adr/. Use when the user wants to improve architecture, find refactoring opportunities, consolidate tightly-coupled modules, or make a codebase more testable and AI-navigable.

0 installs
▌
PDF · davie521 bundle
PDF Processing Guide

0 installs
▌
API Design · davie521
REST API design patterns including resource naming, status codes, pagination, filtering, error responses, versioning, and rate limiting for production APIs. Use when designing or reviewing REST endpoints, or deciding between status codes, pagination styles, or versioning strategies.

0 installs
▌
Critique · davie521 bundle
Evaluate design from a UX perspective, assessing visual hierarchy, information architecture, emotional resonance, cognitive load, and overall quality with quantitative scoring, persona-based testing, and actionable feedback. Use when the user asks to review, critique, evaluate, 评审, or give feedback on a design or component. Diagnosis only — critique judges whether the experience works; to produce or modify a design, use ui-ux-pro-max.

0 installs
▌
Bun Runtime · davie521
Bun as runtime, package manager, bundler, and test runner. When to choose Bun vs Node, migration notes, and Vercel support.

0 installs
▌
Cpr · davie521
Automated Git PR workflow: detect status → commit → create PR → wait for CI → fix failures → review Copilot comments critically → loop until all green → auto-merge. Use when user says 'cpr' or 'cl', wants to push code through the full PR pipeline, or wants to review and fix Copilot code review comments.

0 installs
▌
Research · davie521
Multi-source deep research. Prefers the exa MCP for search (firecrawl for scraping), falls back to the bundled /deep-research workflow or built-in WebSearch when they are absent. Searches the web, synthesizes findings, and delivers cited reports with source attribution. Use when the user wants thorough research on any topic with evidence and citations.

0 installs
▌
Feishu · davie521
Feishu (Lark) MCP integration: send messages, create groups, import/read documents, query users. No file upload/download and no direct cloud-doc editing (official lark-mcp limitations). Use when user mentions '飞书'/Feishu/Lark, or asks to send a message to a Feishu group or user, create a Feishu group, import or read a Feishu doc, or look up Feishu contacts. Requires the 'feishu' MCP server (official lark-mcp CLI, tools like mcp__feishu__im_v1_*) to be configured — if its tools are absent, tell the user to configure it instead of guessing.

0 installs
▌
Crosspost · davie521
Multi-platform content adaptation across X, LinkedIn, Threads, Bluesky, TikTok, YouTube, and newsletters. Drafts a platform-native version for each target platform — it does not post or publish anything. Never produces identical copy cross-platform. Use when the user wants to distribute or repurpose content across platforms.

0 installs
▌
Codex Review · davie521
Default code review route — runs Codex (read-only `codex exec` via the codex-run wrapper) with the multi-language code-review methodology (severity matrix, file:line, scope triage, mandatory security pass). Read-only, no edits. Trigger when the user says "code review" / "审一下" / "review 一下" / "审这个 diff" / "用 codex 审" / "second opinion" / "复审" / "看看这次改动" or asks for any code review on a diff, file, or PR. This is the preferred entry point — route to the built-in /code-review skill when the user wants inline PR comments (--comment) or the findings auto-applied (--fix), and to the Claude-subagent dispatcher (code-review:code-review) only when the user explicitly asks for "Claude reviewers" / "subagent review" / "the multi-agent code-review".

0 installs
▌
To Prd · davie521
Turn the current conversation context into a PRD, saved as a local markdown file by default (published to the project issue tracker only when the user asks). Use when user wants to create a PRD from the current context.

0 installs
▌
Triage · davie521 bundle
Triage issues through a state machine driven by triage roles. Use when user wants to create an issue, triage issues, review incoming bugs or feature requests, prepare issues for an AFK agent, or manage issue workflow.

0 installs
▌
Cheatsheet · davie521 bundle
Create dense, printable A4 landscape cheatsheets from Markdown using a custom syntax with multi-column layout. Use when the user asks to create a cheatsheet, revision sheet, crib sheet, or condensed reference document for exams, courses, or technical topics. Supports math formulas, tables, highlights, notes, and multi-page output. Converts Markdown to styled HTML and exports to PDF for printing.

0 installs
▌
E2e Testing · davie521
Playwright E2E testing patterns, Page Object Model, configuration, CI/CD integration, artifact management, and flaky test strategies.

0 installs
▌
Skill Comply · davie521 bundle
Visualize whether skills, rules, and agent definitions are actually followed — auto-generates scenarios at 3 prompt strictness levels, runs agents, classifies behavioral sequences, and reports compliance rates with full tool call timelines

0 installs
▌
Docker Patterns · davie521
Docker and Docker Compose patterns for local development, container security, networking, volume strategies, and multi-service orchestration. Use when writing Dockerfiles or compose files, hardening images, or debugging container networking and volume issues.

0 installs
▌
UI UX Pro Max · davie521 bundle
UI/UX design intelligence. 67 styles, 96 palettes, 57 font pairings, 25 charts, 13 stacks (React, Next.js, Vue, Svelte, SwiftUI, React Native, Flutter, Tailwind, shadcn/ui). Actions: plan, build, create, design, implement, review, fix, improve, optimize, enhance, refactor, check UI/UX code. Projects: website, landing page, dashboard, admin panel, e-commerce, SaaS, portfolio, blog, mobile app, .html, .tsx, .vue, .svelte. Elements: button, modal, navbar, sidebar, card, table, form, chart. Styles: glassmorphism, claymorphism, minimalism, brutalism, neumorphism, bento grid, dark mode, responsive, skeuomorphism, flat design. Topics: color palette, accessibility, animation, layout, typography, font pairing, spacing, hover, shadow, gradient. Integrations: shadcn/ui MCP for component search and examples.

0 installs
▌
Caveman · davie521
Ultra-compressed communication mode. Substantially cuts token usage by dropping filler, articles, and pleasantries while keeping full technical accuracy. Designed for English output; limited effect on Chinese conversations. TRIGGER only on explicit invocation: user says "caveman mode", "talk like caveman", "use caveman", or invokes /caveman. DO NOT TRIGGER when user merely asks for a short or concise answer — that is a one-off style request, not a switch into this persistent mode.

0 installs

49 published skills