[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M7EM7D2EFJKCMEPDGWKA9M12":3,"skill-same:01M7EM7D2EFJKCMEPDGWKA9M12":47},{"skillNo":4,"name":5,"nameZh":6,"description":7,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":16,"categories":17,"keywords":18,"publishedAt":19,"linkOnly":20,"markdownContent":21,"markdownBody":22,"frontMatter":23,"refName":25,"commitSha":26,"skillDirectory":27,"forksCount":28,"repositoryHtmlUrl":29,"repositoryHomepageUrl":6,"ownerHtmlUrl":30,"ownerBlogUrl":31,"ownerDescription":6,"updatedAt":32,"install":33,"resources":46},"01M7EM7D2EFJKCMEPDGWKA9M12","agent-orchestration",null,"Orchestrate agents at runtime — task routing, scheduling, parallel execution, result aggregation, and supervision dashboards. Use when running agents as a managed fleet rather than one-off calls.","TanNhatCMS\u002Fjarvis","TanNhatCMS","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F113796420?v=4","01M753MV17KG5CD3152G3WQEQY","KietNT","01M7EKEDYVYQ6PWVJJMZ911VMX",0,"MIT","https:\u002F\u002Fgithub.com\u002FTanNhatCMS\u002Fjarvis\u002Fblob\u002Fmain\u002Fskills\u002Fams-agent-orchestration\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\nname: agent-orchestration\ndescription: Orchestrate agents at runtime — task routing, scheduling, parallel execution, result aggregation, and supervision dashboards. Use when running agents as a managed fleet rather than one-off calls.\ncategory: ai-research\n---\n\n# Agent Orchestration\n\nOrchestration is the runtime layer: taking agents as workers and running them as a system — \nrouting tasks, scheduling execution, handling parallelism, aggregating results, and supervising the \nwhole fleet. Architecture designs the team; orchestration runs it.\n\n## Overview\n\nAn orchestrator owns the task lifecycle: intake (validate and normalize requests), planning \n(decompose into subtasks), dispatch (assign to agents with briefs), supervision (track progress, \nhandle stalls), aggregation (combine results), and delivery. Underneath: queues, worker pools, \nretries, timeouts, and rate limits. The orchestrator is also the policy enforcement point — \nbudgets, permissions, and escalation all live here.\n\n## When to use\n\n- Running many agent tasks concurrently: batch jobs, user-facing agent fleets, background \nprocessing.\n- Workflows where tasks have dependencies and need scheduling, not just parallel blasting.\n- Production agent systems needing reliability: retries, failover, backpressure.\n- When you need visibility: what's running, what's stuck, what's costing money.\n\n## Core concepts\n\n- **Task model**: tasks as structured objects — id, type, payload, priority, deadline, retry \npolicy. Everything the orchestrator touches is a task.\n- **Routing**: matching tasks to agents by capability, load, and cost. Simple round-robin to start; \ncapability-aware routing as the fleet diversifies.\n- **Scheduling**: ordering with dependencies (DAGs), priorities, and deadlines. Not everything runs \nnow; the scheduler decides what runs when.\n- **Worker pools**: agents as managed workers with concurrency limits, health checks, and warm\u002Fcold \nlifecycle. Protects against overload.\n- **Supervision**: heartbeats, progress reports, stall detection, and intervention — pause, \nreassign, escalate. The orchestrator never fire-and-forgets.\n- **Aggregation**: combining parallel results — merge, vote, synthesize — with conflict \nresolution rules defined up front.\n\n## Practical workflow\n\n1. Define the task schema and the agent capability registry (what each agent\u002Fworker type can do).\n2. Build intake: validate tasks, assign priorities and deadlines, enqueue durably.\n3. Implement dispatch with concurrency limits per worker type; enforce timeouts and retry policies.\n4. Add supervision: heartbeat monitoring, stall detection, and an operator dashboard showing task \nstates.\n5. Implement aggregation for parallel tasks: merge strategy + conflict rules, tested on fixture \ndata.\n6. Load-test: saturate the system, watch backpressure and degradation — graceful slowdown, not \ncollapse.\n\n```text\nOrchestrator components:\nINTAKE → validate, prioritize, enqueue (durable queue)\nSCHEDULER → DAG + priority + deadline ordering\nDISPATCH → capability-matched assignment, concurrency caps\nSUPERVISE → heartbeats, stall detection, intervention\nAGGREGATE → merge\u002Fvote\u002Fsynthesize with conflict rules\nPOLICY → budgets, permissions, escalation (enforced here)\n```\n\n## Common pitfalls\n\n- **No backpressure**: accepting unlimited tasks until the system melts. Queue, shed load, or \ndegrade gracefully — pick one.\n- **Fire-and-forget dispatch**: tasks sent with no tracking. Every dispatch gets a tracked \nlifecycle.\n- **Retry storms**: blind retries amplifying a downstream failure. Exponential backoff, circuit \nbreakers, dead-letter queues.\n- **Invisibility**: no dashboard, no alerts. You find out about failures from users. Instrument \nfrom day one.\n- **Policy in agents**: budgets and permissions implemented per-agent instead of at the \norchestrator. Centralize enforcement.\n- **Aggregation as an afterthought**: parallel results that don't combine cleanly. Design the merge \nbefore parallelizing.\n","# Agent Orchestration\n\nOrchestration is the runtime layer: taking agents as workers and running them as a system — \nrouting tasks, scheduling execution, handling parallelism, aggregating results, and supervising the \nwhole fleet. Architecture designs the team; orchestration runs it.\n\n## Overview\n\nAn orchestrator owns the task lifecycle: intake (validate and normalize requests), planning \n(decompose into subtasks), dispatch (assign to agents with briefs), supervision (track progress, \nhandle stalls), aggregation (combine results), and delivery. Underneath: queues, worker pools, \nretries, timeouts, and rate limits. The orchestrator is also the policy enforcement point — \nbudgets, permissions, and escalation all live here.\n\n## When to use\n\n- Running many agent tasks concurrently: batch jobs, user-facing agent fleets, background \nprocessing.\n- Workflows where tasks have dependencies and need scheduling, not just parallel blasting.\n- Production agent systems needing reliability: retries, failover, backpressure.\n- When you need visibility: what's running, what's stuck, what's costing money.\n\n## Core concepts\n\n- **Task model**: tasks as structured objects — id, type, payload, priority, deadline, retry \npolicy. Everything the orchestrator touches is a task.\n- **Routing**: matching tasks to agents by capability, load, and cost. Simple round-robin to start; \ncapability-aware routing as the fleet diversifies.\n- **Scheduling**: ordering with dependencies (DAGs), priorities, and deadlines. Not everything runs \nnow; the scheduler decides what runs when.\n- **Worker pools**: agents as managed workers with concurrency limits, health checks, and warm\u002Fcold \nlifecycle. Protects against overload.\n- **Supervision**: heartbeats, progress reports, stall detection, and intervention — pause, \nreassign, escalate. The orchestrator never fire-and-forgets.\n- **Aggregation**: combining parallel results — merge, vote, synthesize — with conflict \nresolution rules defined up front.\n\n## Practical workflow\n\n1. Define the task schema and the agent capability registry (what each agent\u002Fworker type can do).\n2. Build intake: validate tasks, assign priorities and deadlines, enqueue durably.\n3. Implement dispatch with concurrency limits per worker type; enforce timeouts and retry policies.\n4. Add supervision: heartbeat monitoring, stall detection, and an operator dashboard showing task \nstates.\n5. Implement aggregation for parallel tasks: merge strategy + conflict rules, tested on fixture \ndata.\n6. Load-test: saturate the system, watch backpressure and degradation — graceful slowdown, not \ncollapse.\n\n```text\nOrchestrator components:\nINTAKE → validate, prioritize, enqueue (durable queue)\nSCHEDULER → DAG + priority + deadline ordering\nDISPATCH → capability-matched assignment, concurrency caps\nSUPERVISE → heartbeats, stall detection, intervention\nAGGREGATE → merge\u002Fvote\u002Fsynthesize with conflict rules\nPOLICY → budgets, permissions, escalation (enforced here)\n```\n\n## Common pitfalls\n\n- **No backpressure**: accepting unlimited tasks until the system melts. Queue, shed load, or \ndegrade gracefully — pick one.\n- **Fire-and-forget dispatch**: tasks sent with no tracking. Every dispatch gets a tracked \nlifecycle.\n- **Retry storms**: blind retries amplifying a downstream failure. Exponential backoff, circuit \nbreakers, dead-letter queues.\n- **Invisibility**: no dashboard, no alerts. You find out about failures from users. Instrument \nfrom day one.\n- **Policy in agents**: budgets and permissions implemented per-agent instead of at the \norchestrator. Centralize enforcement.\n- **Aggregation as an afterthought**: parallel results that don't combine cleanly. Design the merge \nbefore parallelizing.\n",{"name":5,"category":24,"description":7},"ai-research","main","af388a77c2fe5e0b4fd879cca4300a0ffa9a6579","skills\u002Fams-agent-orchestration",1,"https:\u002F\u002Fgithub.com\u002FTanNhatCMS\u002Fjarvis","https:\u002F\u002Fgithub.com\u002FTanNhatCMS","https:\u002F\u002Fmrkiet.name.vn","2026-11-07 14:23:08",{"source":8,"skillName":5,"commands":34,"prompt":44,"zipAvailable":45},[35,38,41],{"tool":36,"command":37},"npx","npx skills add TanNhatCMS\u002Fjarvis --skill agent-orchestration",{"tool":39,"command":40},"bunx","bunx skills add TanNhatCMS\u002Fjarvis --skill agent-orchestration",{"tool":42,"command":43},"pnpm","pnpm dlx skills add TanNhatCMS\u002Fjarvis --skill agent-orchestration","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002FTanNhatCMS\u002Fjarvis\u002Fblob\u002Fmain\u002Fskills\u002Fams-agent-orchestration\u002FSKILL.md",true,[],[48,55,63,70,77,84],{"skillNo":49,"name":50,"nameZh":6,"description":51,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":52,"categories":53,"keywords":54,"publishedAt":19},"01M7EWYJR0J8DARY7E59Z6Y9MV","web-scrape","Turn web pages into LLM-ready markdown via the Firecrawl API (renders JS, anti-bot handling). Use when browser_open\u002Ffetch fails on JS-heavy pages, or when you need clean markdown\u002Fstructured extraction from a site. Requires FIRECRAWL_API_KEY.","https:\u002F\u002Fgithub.com\u002FTanNhatCMS\u002Fjarvis\u002Fblob\u002Fmain\u002Fskills\u002Fweb-scrape\u002FSKILL.md",[],[],{"skillNo":56,"name":57,"nameZh":6,"description":58,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":59,"sourceHtmlUrl":60,"categories":61,"keywords":62,"publishedAt":19},"01M7EWYBN5P6F2AJG3FR24T9BP","design-taste-frontend","tasteskill v2, the anti-slop frontend skill. Art direction for websites, distilled from a study of 247 curated gallery sites (One Page Love, Awwwards Sites of the Day). Deals researched directions (idea, composition, type, palette, photographs, section blueprints) from a library, enforces a strict no-slop house style with a real-browser inspector, and runs a look-and-fix loop on screenshots. Use this whenever you build, design, redesign, restyle, polish or review a landing page, marketing site, portfolio, product page, one-pager, hero or any public-facing web page, in any stack (HTML\u002FCSS, React, Next.js, Vue, Svelte, Astro, Tailwind) — also when the user says a page looks generic, templated, \"AI-made\", sloppy, or asks to make it look designed, premium, award-level or less like every other site. Not for dashboards or app screens.","https:\u002F\u002Fgithub.com\u002FTanNhatCMS\u002Fjarvis\u002Fblob\u002Fmain\u002Fskills\u002Ftaste-design\u002FLICENSE","https:\u002F\u002Fgithub.com\u002FTanNhatCMS\u002Fjarvis\u002Fblob\u002Fmain\u002Fskills\u002Ftaste-design\u002FSKILL.md",[],[],{"skillNo":64,"name":65,"nameZh":6,"description":66,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":67,"categories":68,"keywords":69,"publishedAt":19},"01M7EWY8FQFNQ2GB7NZZWGSSDX","test-driven-development","Use when implementing any feature or bugfix, before writing implementation code","https:\u002F\u002Fgithub.com\u002FTanNhatCMS\u002Fjarvis\u002Fblob\u002Fmain\u002Fskills\u002Fsuperpowers-tdd\u002FSKILL.md",[],[],{"skillNo":71,"name":72,"nameZh":6,"description":73,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":74,"categories":75,"keywords":76,"publishedAt":19},"01M7EWY5SHBFRH1HDHVWM9K7TV","systematic-debugging","Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes","https:\u002F\u002Fgithub.com\u002FTanNhatCMS\u002Fjarvis\u002Fblob\u002Fmain\u002Fskills\u002Fsuperpowers-debugging\u002FSKILL.md",[],[],{"skillNo":78,"name":79,"nameZh":6,"description":80,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":81,"categories":82,"keywords":83,"publishedAt":19},"01M7EWY3NM0Z0JAKS7HDZ24K66","requesting-code-review","Use when completing tasks, implementing major features, or before merging to verify work meets requirements","https:\u002F\u002Fgithub.com\u002FTanNhatCMS\u002Fjarvis\u002Fblob\u002Fmain\u002Fskills\u002Fsuperpowers-code-review\u002FSKILL.md",[],[],{"skillNo":85,"name":86,"nameZh":6,"description":87,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":88,"categories":89,"keywords":90,"publishedAt":19},"01M7EWY0WRMWX4M9ZFY0WKC593","brainstorming","You MUST use this before any creative work - creating features, building components, adding functionality, modifying behavior, or planning anything new, in software or out of it (a talk, a business, a renovation).","https:\u002F\u002Fgithub.com\u002FTanNhatCMS\u002Fjarvis\u002Fblob\u002Fmain\u002Fskills\u002Fsuperpowers-brainstorming\u002FSKILL.md",[],[]]