платформы оркестрации выполнения агентов | AI Agent Skills
Подборка installable AI Agent Skills для developer workflow в orchestration platforms execution для Claude Code, Cursor и Windsurf.
Эта страница посвящена платформам оркестрации, которые координируют сложное выполнение агентов в реальных delivery-пайплайнах. Инструменты поддерживают порядок зависимостей, параллельные задачи, повторы, согласования и наблюдаемость долгих задач, чтобы безопасно управлять многошаговой автоматизацией в продакшене. MCP может быть доступен как интеграция, но главный приоритет — надежность оркестрации.
Primary Install Bridge
Pick One Skill, Then Take the Install Path
This collection should not trap users in comparison mode or pretend to install the whole collection. Its job is to narrow the shortlist to one skill, then send the next click into installation, validation, and rollout. Installation happens on the skill path.
The Next Click Should Keep Narrowing, Not Reset Back To A Generic Directory
Once the install path is clear, move into the solution, CLI, or editorial surface that best matches this collection. That keeps platform, framework, and operations demand narrowing into a more verifiable high-intent journey.
Reviewed on 2026-04-17 against orchestration fit, installation clarity, approval-aware execution, and production handoff value. This page is now positioned as an install-first orchestration entry point instead of a generic platform roundup.
We prioritize this page because orchestration-intent users usually need one platform they can install, validate, and prove inside a real multi-step workflow before standardizing anything broader.
Trust Signals
- Entries are chosen for practical orchestration value such as routing, approvals, retries, visibility, and long-running execution support.
- Selection favors tools with public docs and setup paths that teams can validate before pushing them into production workflows.
- The page is curated for operator-facing execution reliability, not for abstract platform adjacency or vague agent branding.
Grouping Logic
- Lead with tools that can own multi-step execution without hiding approval boundaries or retry behavior.
- Keep the shortlist compact enough for fast comparison while still covering orchestration, handoff, and production visibility.
- Use installation as the bridge from orchestration comparison into validated workflow execution.
Maintenance & Review
Last Reviewed
2026-04-17
Cadence
Re-check when install flow, maintainer posture, or orchestration relevance changes upstream; otherwise review monthly.
Maintained By
Killer-Skills editorial review within the recovery-focused authority queue.
Verification
Validate installability, approval clarity, execution fit, and maintainer trust before retaining or adding an entry.
Execution Examples
How These Skills Work Together In Practice
Validate one approval-gated workflow lane first
Use this page when you need one orchestration layer that can move a real workflow through routing, approvals, retries, and handoff without turning selection into another architecture debate.
1. Open the installation docs before comparing more orchestration platforms.
2. Choose one platform that best matches your approval boundaries and execution model.
3. Install it and verify the CLI write path, retry behavior, and first operator checkpoint.
4. Only after the base lane works, expand it to broader workflow automation.
Prove retry and handoff behavior before scaling
Treat the collection as an editorial filter when you need orchestration tools that can survive long-running jobs and operator checkpoints without hiding failure modes.
1. Check whether the platform has stable ownership and visible install guidance.
2. Review operator behavior so the team knows where retries, approvals, and logs actually live.
3. Use one validated workflow lane before standardizing orchestration across more systems.
4. Document the chosen orchestration baseline after the first clean production-style test.
The component-refactoring skill simplifies high-complexity React components, improving code quality and reducing maintenance efforts for developers. It utilizes patterns and workflows to refactor components, making them more efficient and easier to understand.
Generate Vitest + React Testing Library tests for Dify frontend components, hooks, and utilities. Triggers on testing, spec files, coverage, Vitest, RTL, unit tests, integration tests, or write/review test requests.
Guide for implementing oRPC contract-first API patterns in Dify frontend. Trigger when creating or updating contracts in web/contract, wiring router composition, integrating TanStack Query with typed contracts, migrating legacy service calls to oRPC, or deciding whether to call queryOptions directly vs extracting a helper or use-* hook in web/service.
This frontend-code-review skill reviews frontend files, such as .tsx, .ts, and .js, and applies a checklist to ensure code quality, performance, and business logic, helping developers and teams maintain high standards.
Design and implement the Guardian AI companion evolution system - from Level 1 Spark to Level 50 Transcendent. XP mechanics, personality adaptation, and player progression.
/Lumina-Intelligence, /Luminor-intelligence & /StarlightOrchestrator building Arcanea with swarm intelligence that overlays any AI. Ten Gates, Guardians & Godbeasts, context that compounds, and the creative civilization OS for creators who build universes.
Implement ReasoningBank adaptive learning with AgentDBs 150x faster vector database. Includes trajectory tracking, verdict judgment, memory distillation, and pattern recognition. Use when building self-learning agents, optimizing decision-making, or implementing experience replay systems.
Apply and enforce brand voice across all content creation. Manages voice attributes, tone adaptation by channel, style rules, and terminology. Use when writing content, reviewing drafts, or defining brand voice for a new project. Loads voice config from CREATOR.md or creator-memory/voice.md.
If This Page Is Close, Keep Narrowing With Adjacent Authority Pages
Do not reset back to the generic directory. Move sideways through these adjacent high-intent collections to narrow the shortlist toward the install path that best matches your team.
This collection should not keep users browsing forever. These three questions explain how to shortlist, install, and validate the next step.
Для каких рабочих процессов созданы эти подборки?
Эти подборки построены вокруг автоматизации рабочих процессов, документов, данных и переиспользуемых стеков навыков.
Чем подборка отличается от основного каталога навыков?
Каталог лучше подходит для прямого поиска, а подборки помогают найти взаимодополняющие навыки, объединённые вокруг рабочего процесса.
Можно ли установить эти подборки для Claude Code или Cursor?
Да. Навыки в этих подборках обычно работают в Claude Code, Cursor, Windsurf и других поддерживаемых средах через единый процесс установки.
Additional Recovery Paths
Use These Additional Paths If You Need One More Step To Narrow The Decision
These are the supporting surfaces for this collection after the install direction is clear and the primary next paths have already narrowed the decision.
