int2t05
User48 engineering skills for Claude Code across the SDLC — shared principles injected at session start. Works with Codex, Cursor, Cline & other markdown agents.
Categories
Indexed Skills (47)
handoff
Use when handing off work to another session or agent — produces a structured handoff brief capturing context, decisions, and next steps. Triggers on "handoff", "hand over", "交接", "移交工作".
brainstorm
Use before creative work, or when the ask is underspecified — one-question-at-a-time dialogue that sharpens a vague idea into a concrete proposal. Triggers on "brainstorm", "grill me", "interview me", "refine this idea", "头脑风暴", "帮我打磨想法", "盘问我".
oss-strategy
Use when deciding open source strategy — business model, COSS, open core, commercialization, or growth. Triggers on "open source strategy", "open source to paid", "open source business model", "OSS 策略", "DevHunt", "开源策略", "开源商业模式".
spec
Use when starting a new project, feature, or significant change — writes a spec/PRD covering objectives, structure, commands, code style, testing, and boundaries before any code. Triggers on "write spec", "create prd", "spec out", "写需求文档", "写规格", "需求文档" — also when user says "要做什么" / "需求是什么".
codebase-design
Use when designing deep modules, finding refactoring or deepening opportunities, or making a codebase more testable and AI-navigable. Presents deepening opportunities and works through the one you pick. Triggers on "deep modules", "refactor architecture", "deepening", "深化模块", "重构架构", "代码库设计". Not for executing the change (use refactoring), single-file clarity (use simplify), or greenfield architecture (use architecture) — this skill designs and audits, it does not edit code.
frontend-design
Use when building web components, pages, or applications with distinctive, production-grade design quality — color, typography, layout, interaction states, and platform conventions. Triggers on "frontend", "UI design", "design", "组件设计", "界面设计", "前端设计", "视觉风格".
breakdown
Use when breaking a plan, spec, or conversation into tracer-bullet tickets, each declaring its blocking edges — for work too large for one session, builds a shared map of decision tickets resolved one at a time. Triggers on "break into tickets", "decompose", "wayfinder", "拆解任务", "拆票", "任务��解".
implement
Use when implementing the work described by a spec or tickets — drives TDD within each slice, runs typechecks and tests regularly, and closes with code-review before committing. Also covers lightweight changes (small edits, mechanical renames, project scaffolding) that don't need a full spec. Triggers on "implement", "build this", "code the feature", "实现", "编码", "改这个配置", "重命名", "搭项目骨架" — also when user says "改一下这个" / "加个功能". Not for pure test-driven exploration with no spec (use tdd).
simplify
Use when the code is too complex — clarity over cleverness, removes speculative abstractions, dead complexity, and earns-its-cost structures. Triggers on "simplify", "too complex", "refactor for clarity", "简化", "太复杂", "重构求清晰". Not for cross-module structural changes (use refactoring) or architecture audit (use codebase-design) — single-file clarity stays here.
e2e-testing
Use when writing end-to-end or browser tests — user journeys, form submission, and runtime UI verification through a browser automation tool (Playwright by default, or your framework's equivalent). Triggers on "playwright", "e2e test", "browser test", "end-to-end", "端到端测试", "浏览器测试".
test-generation
Use when generating test files for existing code or from a spec — language-agnostic, works with any test framework. Triggers on "generate tests", "write tests for", "生成测试", "生成测试代码", "补测试".
oss-polish
Use when polishing an open source project's GitHub presence — README, topics/About description, commit-history narrative story, and trending-repo positioning. Triggers on "polish my repo", "beautify open source project", "开源项目美化", "优化项目展示".
observability
Use when adding logs, metrics, alerts, or instrumentation to a system — making runtime behavior observable and debuggable in production. Triggers on "add logging", "metrics", "alerting", "instrumentation", "加日志", "可观测性", "��控告警".
using-skills
Use when starting a session or deciding which skill applies — maps incoming work to the right skill across the 9 SDLC phases. Triggers on "which skill should I use", "route this task", "用哪个技能", "路由".
api-testing
Use when testing APIs or designing API test strategies — contract testing, REST/GraphQL testing, and integration testing. Triggers on "test API", "contract testing", "integration test", "API 测试", "接口测试", "契约测试".
tdd
Use when implementing any feature or bugfix, before writing implementation code — red-green-refactor loop, one vertical slice at a time. Triggers on "tdd", "test-driven", "red green refactor", "测试驱动开发", "红绿重构", "测试驱动". Not for spec-driven slice implementation (use implement — it drives TDD within each slice) or generating tests for existing code (use test-generation).
ci-cd
Use when working on CI/CD pipelines and automation — build, test, and deploy automation, pipeline design, and deployment strategies. Triggers on "CI pipeline", "GitHub Actions", "deployment strategy", "CI/CD", "流水线", "持续集成", "部署策略".
git-workflow
Use when committing, branching, resolving merge or rebase conflicts, or setting up git guardrails and pre-commit hooks — resolves conflicts by intent traced to each side's source, never --abort. Triggers on "commit", "merge conflict", "rebase", "pre-commit", "提交", "合并冲突", "分支管理".
shipping
Use when deploying or launching to production — checklist-driven launch with rollback readiness and launch-day verification. Triggers on "ship", "deploy", "launch", "go live", "上线", "发布", "部署到生产".
code-review
Use when reviewing code — pre-merge two-axis review (Standards + Spec), in-flight adversarial doubt, evidence-before-done verification, or receiving review feedback. Triggers on "review this", "code review", "before merge", "代码审查", "合并前审查", "帮我看看这代码", "能合并吗", "are you sure", "verify before done", "收到评审意见". Not for machine-detectable lint (use linting) or runtime bugs (use debugging).
design-research
Use when retrieving real design references — landing pages, user flows, visual styles — from free public galleries to ground imagegen or frontend-design work in what real products actually do. Triggers on "design references", "real UI references", "find design inspiration", "设计参考", "真实设计参考", "找设计灵感", "设计灵感". Not for generating design images (use imagegen) or writing frontend code (use frontend-design).
imagegen
Use when generating premium design reference images across three modes — brand identity boards, website section images, or mobile app screens — art-directed and implementation-friendly, not generic AI slop. Output is images only, never code. Triggers on "brand kit", "logo 设计", "品牌系统", "web design image", "landing page image", "网站设计图", "mobile design image", "app screen image", "移动端设计图", "手机界面图". Not for frontend code (use frontend-design) or image-to-code pipeline (use image-to-code).
multi-agent-orchestration
Use when executing work through subagents — dispatching independent tasks to parallel isolated-context agents, or driving a plan as a sequence of implementer-subagents with review checkpoints. Scales beyond a single context window. Triggers on "parallel agents", "subagent execution", "dispatch tasks", "fan out", "并行 agent", "子代理执行", "派发任务". Not for breaking a plan into tickets (use breakdown) or doing the implementation yourself (use implement).
refactoring
Use when restructuring code without changing behavior — extract modules, move classes, redraw dependency boundaries, split or merge files. Behavior-preserving structural changes that make the codebase easier to evolve, distinct from single-file clarity cleanup. Triggers on "refactor structure", "extract module", "move class", "split file", "change dependency", "重构结构", "提取模块", "拆分文件", "改依赖图". Not for single-file clarity cleanup (use simplify), architecture audit producing a design doc (use codebase-design), or lint/style fixes (use linting).
a11y-review
Use when reviewing changes for accessibility — WCAG conformance, axe-core findings, keyboard and screen-reader support, contrast. Triggers on "accessibility", "a11y", "WCAG", "screen reader", "无障碍", "可访问性". Not for general code quality (use code-review), machine-detectable lint (use linting), or runtime behavior bugs (use debugging).
linting
Use when fixing machine-detectable code issues — run linters and static analyzers (ESLint, Ruff, Clippy, golangci-lint), resolve findings, suppress false positives with justification, tighten config. Automated quality gate, distinct from human-judgment review. Triggers on "lint", "linting", "fix eslint", "static analysis", "lint 修复", "静态分析", "格式检查", "代码规范检查". Not for human-judgment code review (use code-review), runtime behavior bugs (use debugging), or behavior-preserving structural refactoring (use refactoring).
skill-authoring
Use when creating, editing, or evaluating a skill in this collection — writes SKILL.md frontmatter and four sections, generates the Codex adapter, runs the validator, and tunes description triggering. The authoring half of using-skills (which routes/consumes). Triggers on "write a skill", "edit skill", "skill eval", "tune triggers", "写技能", "改技能", "技能评估". Not for routing to an existing skill (use using-skills) or building an MCP server (use mcp-builder if added).
api-design
Use when designing APIs or interfaces — REST/GraphQL contracts, request/response shapes, versioning, error models, and interface ergonomics. Triggers on "design API", "REST contract", "GraphQL schema", "接口设计", "API 契约", "API 设计".
architecture
Use when designing high-level system architecture, reviewing existing designs, or making architectural decisions — produces ADRs, architecture diagrams, and evaluates scalability/NFR trade-offs. Triggers on "system design", "架构设计", "ADR", "scalability", "系统设计", "架构决策".
security-review
Use when reviewing changes for security — secrets, auth, injection, access control, and hardening. Triggers on "security review", "check for vulnerabilities", "安全审查", "安全审计", "漏洞检查".
deprecation-migration
Use when deprecating old code or APIs, migrating to a new system, or upgrading dependencies — staged deprecation paths and migration strategies that preserve behavior across the transition. Triggers on "deprecate", "migrate", "sunset API", "���用", "迁移", "升级依赖", "更新这个库", "upgrade dependency".
documentation-audit
Use when documentation drifts from code — syncs PRD/TECH/API/FLOW/TODO to the codebase and produces an audit report. Triggers on "docs out of date", "documentation drift", "sync docs", "文档同步", "写文档", "写 README", "文档化这个功能", "文档审计".
schema-design
Use when designing a data model for a new feature or bounded context — entities, relationships, normalization, indexing, constraints, partitioning. Triggers on "data model", "schema design", "database design", "ER model", "数据模型", "表结构设计", "数据库设计".
i18n
Use when internationalizing an application — message extraction, ICU/MessageFormat, locale routing, RTL layout, pluralization/gender, and locale-aware formatting. Triggers on "i18n", "localization", "l10n", "RTL", "国际化", "本地化", "多语言".
auth-implementation
Use when implementing authentication — login, sessions, JWT, OAuth2/OIDC, password hashing, and RBAC/ABAC authorization. Language-agnostic patterns for building auth correctly the first time. Triggers on "implement auth", "password hashing", "OAuth integration", "实现认证", "实现登录", "OAuth 集成", "token 轮换". Not for auditing existing auth for vulnerabilities (use security-review) or designing auth architecture at system level (use architecture).
error-handling
Use when designing error-handling strategy for a feature — error classification, throw-vs-return conventions, retry/circuit-breaker/fallback decisions, and error-to-user-message mapping. Triggers on "error handling", "retry strategy", "circuit breaker", "error propagation", "错误处理", "重试策略", "熔断", "降级" — also when user says "这个错误该怎么处理" / "要不要重试". Not for diagnosing a specific bug (use debugging); system-level resilience architecture (use architecture).
purity-audit
Use when auditing whether code or docs read as first-written — flags tracing residue, decorative structure, dead content, doc/impl gaps, and language/platform residue against the 16-point purity checklist; produces an audit report. Triggers on "purity audit", "纯净审计", "残留审计", "洁净化" — also when user says "这些文件像改了好几遍" / "清掉历史残留". Not for syncing formal docs to code behavior (use documentation-audit) or reviewing a diff for bugs (use code-review).
market-research
Use when the user wants market sizing, competitor comparisons, investor due diligence, industry intelligence, fund research, or technology scans that inform business decisions. Triggers on "market sizing", "competitor analysis", "due diligence", "市场调研", "市场规模", "竞���分析".
tech-selection
Use when choosing or comparing a technology stack, library, framework, open-source project, or repository for a concrete requirement. Triggers on "技术选型", "方案对比", "选哪个", "tech stack", "library comparison", "技术对比".
domain-modeling
Use when building or sharpening a project's domain model — challenging terms, stress-testing with scenarios, and updating CONTEXT.md and ADRs to establish a shared ubiquitous language. Triggers on "domain model", "ubiquitous language", "CONTEXT.md", "领域模型", "统一语言", "领域建模".
prototype
Use when a design question is best answered by a throwaway prototype — a single shareable HTML file for state/logic, or several toggleable UI variations to compare. Triggers on "prototype", "throwaway", "mockup", "compare layouts", "validate the interaction", "sketch out", "try this quickly", "build a demo", "原型", "试做", "试这个方案", "搭个快速 demo".
context-engineering
Use when the agent needs better context — assembling the right files, definitions, and prior decisions before implementing, or explaining existing code to a user. Triggers on "agent lacks context", "what files matter", "解释这段代码", "这个模块怎么工作", "带我过一遍代码库".
brandkit
brand identity image generation — logo concepts, identity boards, color palettes, typography, mockups for premium brand systems. Triggers on "brand kit", "品牌识别", "logo 设计", "品牌系统", "品牌设计".
imagegen-mobile
Mobile app screen and flow image generation — iOS/Android/cross-platform concepts with phone mockup framing, clean hierarchy, and strong multi-screen consistency. Triggers on "mobile design image", "app screen image", "mobile mockup", "app flow", "移动端设计图", "app 屏幕图", "手机界面图", "移动端流程图".
imagegen-web
Website design reference image generation — one horizontal image per section, premium art direction for landing pages and marketing sites that developers or coding models can recreate. Triggers on "web design image", "website mockup", "landing page image", "section image", "网站设计图", "网页参考图", "落地页配图".
prompt-engineering
Use when designing prompts, evals, or LLM-powered features — prompt architecture, model selection, guardrails, and eval harnesses for prompt-as-product surfaces. Triggers on "prompt engineering", "LLM feature", "eval harness", "prompt design", "提示词工程", "LLM 特性", "prompt 设计".
load-testing
Use when validating capacity under load — generate realistic and adversarial traffic, find breaking points, characterize saturation, validate autoscaling. Triggers on "load test", "stress test", "capacity", "k6", "Locust", "wrk", "压测", "压力测试", "容量测试".
Bio shown is the top-scored skill's repo description as a fallback — real GitHub bios land in a future update.