发现 工具与集成 相关 Skills，比较来源仓库和用途，查看适合你工作流的技能说明。
Multi-agent systems with LangGraph - supervisor/swarm patterns, state coordination, multi-provider routing. Use when building multi-agent workflows, coordinating agents, or need cost-optimized orchestration. Us...
Parallel development with git worktrees and Claude Code agents. Handles Ghostty terminal launching, port allocation, and global registry. Use when creating worktrees, managing parallel development, or launching...
B2B go-to-market strategy, pricing models, ICP development, positioning, and competitive intelligence. Use when planning GTM strategy, setting pricing, defining ICP, or evaluating opportunities.
Business model design using Alexander Osterwalder's 9 building blocks. Use when: business model, canvas, value proposition, customer segments, revenue streams, startup planning, analyze business, business strat...
B2B content marketing - thought leadership, SEO, LinkedIn, blog posts, case studies, and video scripts. Use when creating content strategy, writing posts, or building demand gen assets.
CRM integration patterns for Close CRM, HubSpot, and Salesforce. Use when: Close CRM, HubSpot, Salesforce, CRM API, lead sync, deal sync, activity logging, CRM webhook, pipeline automation, contact enrichment.
Production voice AI agents with sub-500ms latency. Groq LLM, Deepgram STT, Cartesia TTS, Twilio integration. No OpenAI. Use when: voice agent, phone bot, STT, TTS, Deepgram, Cartesia, Twilio, voice AI, speech t...
Stripe integration patterns for Next.js + Supabase - payments, subscriptions, webhooks, credit systems, checkout. Use when: stripe, payments, billing, subscription, webhook, checkout, add payments to project.
Project workflow system - cost tracking, parallel execution, security gates, agent orchestration. Use when: start day, begin session, status check, new feature, build, implement, end day, wrap up, debug, invest...
Maintains project context and progress tracking across Claude sessions. Use at session start to load context, on session end to save progress. Triggers: "load project context", "save context", "end session", "w...
Deploy GPU workloads to RunPod serverless and pods - vLLM endpoints, A100/H100 setup, scale-to-zero, cost optimization. Use when: deploy to RunPod, GPU serverless, vLLM endpoint, scale to zero, A100 deployment,...
Use for all task management - prog is the primary system for tracking work across sessions; use TodoWrite only for quick single-message tasks
Execute a written implementation plan with batch processing and human review checkpoints. Use when you have a plan file ready for implementation in a parallel session.
Create comprehensive implementation plans with bite-sized tasks. Use after brainstorming produces a design, before touching code. Documents exact files, complete code, commands, expected output.
Execute implementation plans in same session using fresh subagent per task + two-stage review (spec compliance, then code quality)
Interview the user to build a detailed specification before implementing a feature. Use when requirements are ambiguous or feature has significant complexity.
Turn ideas into fully formed designs through collaborative dialogue. Use BEFORE any creative work - creating features, building components, adding functionality, or modifying behavior.
Comprehensive pull request review covering code quality, tests, security, and documentation. Use when reviewing PRs or when asked to review changes.
NS 8407:2011 totalentreprisekontrakt. Bruk ved arbeid med KOE, endringsordrer, vederlag eller fristforlengelse.
Statiske analyseverktøy for drift-sjekk, sikkerhetsscan og kodekvalitet. Bruk før commit eller for å finne synkroniseringsproblemer.
Guide til Event Sourcing arkitekturen. Bruk ved arbeid med events, state-projeksjoner, eller forretningsregler.
Tilgjengelighetsverktøy for WCAG-validering. Bruk ved arbeid med farger, kontrast eller UI-komponenter.
Dokumentasjonsvedlikehold og synkroniseringssjekk. Bruk etter større endringer, før release, eller når du mistenker utdatert dokumentasjon.
Local-first, multi-product backlog management with agent collaboration discipline.
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。