[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M7CAF87YVC5MXS8RADB21M8C":3,"skill-same:01M7CAF87YVC5MXS8RADB21M8C":46},{"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":24,"commitSha":25,"skillDirectory":26,"forksCount":14,"repositoryHtmlUrl":27,"repositoryHomepageUrl":6,"ownerHtmlUrl":28,"ownerBlogUrl":29,"ownerDescription":30,"updatedAt":31,"install":32,"resources":45},"01M7CAF87YVC5MXS8RADB21M8C","orchestration-workflow",null,"Guide for working with the multi-agent orchestration system in DEVS. Use this when asked to modify task orchestration, agent coordination, or workflow execution.","shenyimings\u002Fskillet","shenyimings","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F89332218?v=4","01M7C9KHMXPPXMQJCKN78FKB06","Yiming Shen","01M7C9KHP4KS809T2A9T64YDWE",0,"MIT","https:\u002F\u002Fgithub.com\u002Fshenyimings\u002Fskillet\u002Fblob\u002Fmain\u002Fbenchmark\u002Fwild\u002Fdev\u002Fsafe\u002Fcodename-co__devs__.github__skills__orchestration-workflow\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\nname: orchestration-workflow\ndescription: Guide for working with the multi-agent orchestration system in DEVS. Use this when asked to modify task orchestration, agent coordination, or workflow execution.\n---\n\n# Orchestration Workflow for DEVS\n\nThe DEVS orchestration system coordinates multiple AI agents to complete complex tasks autonomously.\n\n## Architecture Overview\n\n```\nUser Request → WorkflowOrchestrator → TaskAnalyzer → Agent Execution → Validation\n                     ↓\n              Strategy Selection\n              (single-pass \u002F multi-pass)\n                     ↓\n              Team Building → Coordinated Execution → Artifact Creation\n```\n\n## Core Components\n\n### WorkflowOrchestrator (`src\u002Flib\u002Forchestrator.ts`)\n\nCentral coordination hub that:\n\n- Prevents duplicate processing via prompt hashing\n- Selects execution strategy based on complexity\n- Manages agent recruitment and team building\n- Handles error recovery and retries\n\n### TaskAnalyzer (`src\u002Flib\u002Ftask-analyzer.ts`)\n\nLLM-powered analysis that:\n\n- Breaks down complex prompts into requirements\n- Assesses task complexity (simple\u002Fcomplex)\n- Identifies required skills\n- Generates subtasks with dependencies\n\n### RequirementValidator (`src\u002Flib\u002Frequirement-validator.ts`)\n\nValidates task deliverables against requirements:\n\n- Functional requirements (feature implementation)\n- Non-functional requirements (performance, usability)\n- Constraints (technology, time, resources)\n\n## Execution Strategies\n\n### Single-Pass Strategy\n\nFor simple tasks that one agent can complete:\n\n```typescript\nasync function executeSinglePass(task: Task, agent: Agent): Promise\u003Cvoid> {\n  \u002F\u002F 1. Execute with enhanced context\n  const result = await executeTaskWithAgent(task, agent, task.description)\n\n  \u002F\u002F 2. Create artifact from result\n  await createArtifact({\n    taskId: task.id,\n    agentId: agent.id,\n    type: 'deliverable',\n    content: result,\n  })\n\n  \u002F\u002F 3. Validate against requirements\n  const validation = await validateRequirements(task, [artifact])\n\n  \u002F\u002F 4. Handle validation failures\n  if (!validation.allSatisfied) {\n    await createRefinementTask(task, validation.failures)\n  }\n}\n```\n\n### Multi-Pass Strategy\n\nFor complex tasks requiring multiple agents:\n\n```typescript\nasync function executeMultiPass(task: Task): Promise\u003Cvoid> {\n  \u002F\u002F 1. Break down into subtasks\n  const subtasks = await taskAnalyzer.breakdown(task)\n\n  \u002F\u002F 2. Build specialized team\n  const team = await buildTeam(subtasks)\n\n  \u002F\u002F 3. Execute with dependency resolution\n  await coordinateTeamExecution(subtasks, team)\n\n  \u002F\u002F 4. Validate all deliverables\n  const validation = await validateAllRequirements(task, subtasks)\n}\n```\n\n## Team Coordination\n\n```typescript\nasync function coordinateTeamExecution(\n  tasks: Task[],\n  team: Agent[],\n): Promise\u003Cvoid> {\n  const executedTasks = new Set\u003Cstring>()\n\n  while (executedTasks.size \u003C tasks.length) {\n    \u002F\u002F Find tasks with satisfied dependencies\n    const readyTasks = tasks.filter(\n      (task) =>\n        !executedTasks.has(task.id) &&\n        task.dependencies.every((depId) => executedTasks.has(depId)),\n    )\n\n    \u002F\u002F Execute in parallel batches\n    const batch = readyTasks.slice(0, team.length)\n    await Promise.all(\n      batch.map((task, index) => {\n        const agent = team[index % team.length]\n        return executeTaskWithAgent(task, agent, task.description)\n      }),\n    )\n\n    batch.forEach((task) => executedTasks.add(task.id))\n  }\n}\n```\n\n## Context Sharing\n\nThe ContextBroker enables inter-agent communication:\n\n```typescript\nimport { ContextBroker } from '@\u002Flib\u002Fcontext-broker'\n\n\u002F\u002F Agent publishes context\nawait ContextBroker.publish({\n  type: 'finding',\n  agentId: agent.id,\n  content: 'Discovered that the API requires authentication',\n  keywords: ['api', 'authentication', 'security'],\n})\n\n\u002F\u002F Another agent retrieves relevant context\nconst relevantContext = await ContextBroker.getRelevant(['api', 'design'])\n```\n\n## Task States\n\n```typescript\ntype TaskStatus = 'pending' | 'in_progress' | 'completed' | 'failed'\n```\n\nState transitions:\n\n- `pending` → `in_progress` (when assigned to agent)\n- `in_progress` → `completed` (when validated successfully)\n- `in_progress` → `failed` (when validation fails after retries)\n- `failed` → `in_progress` (when retry initiated)\n\n## Requirement Types\n\n```typescript\ninterface Requirement {\n  id: string\n  type: 'functional' | 'non-functional' | 'constraint'\n  description: string\n  priority: 'must' | 'should' | 'could' | 'wont'\n  source: 'explicit' | 'implicit' | 'inferred'\n  validationStatus?: 'satisfied' | 'pending' | 'failed'\n  evidence?: string[]\n}\n```\n\n## Creating Custom Orchestration Logic\n\nWhen extending orchestration:\n\n```typescript\nimport { WorkflowOrchestrator } from '@\u002Flib\u002Forchestrator'\nimport { TaskAnalyzer } from '@\u002Flib\u002Ftask-analyzer'\nimport { getAgentById, createAgent } from '@\u002Fstores\u002FagentStore'\n\nasync function customOrchestration(prompt: string): Promise\u003Cvoid> {\n  \u002F\u002F 1. Analyze the task\n  const analysis = await TaskAnalyzer.analyze(prompt)\n\n  \u002F\u002F 2. Find or create suitable agent\n  let agent = await findAgentWithSkills(analysis.requiredSkills)\n  if (!agent) {\n    agent = await createAgent({\n      name: 'Dynamic Agent',\n      role: analysis.suggestedRole,\n      instructions: analysis.suggestedInstructions,\n    })\n  }\n\n  \u002F\u002F 3. Create task with requirements\n  const task = await createTask({\n    title: analysis.title,\n    description: prompt,\n    complexity: analysis.complexity,\n    requirements: analysis.requirements,\n    assignedAgentId: agent.id,\n  })\n\n  \u002F\u002F 4. Execute based on complexity\n  if (analysis.complexity === 'simple') {\n    await executeSinglePass(task, agent)\n  } else {\n    await executeMultiPass(task)\n  }\n}\n```\n\n## Error Handling\n\nThe orchestrator implements multi-level error handling:\n\n1. **Orchestration Level**: Duplicate prevention, graceful degradation\n2. **Validation Level**: Automatic refinement tasks on failure\n3. **Agent Level**: Fallback agent creation when recruitment fails\n4. **Network Level**: LLM provider fallbacks, timeout handling\n\n```typescript\ntry {\n  await orchestrator.execute(prompt)\n} catch (error) {\n  if (error instanceof DuplicatePromptError) {\n    \u002F\u002F Already processing this prompt\n    return existingWorkflow\n  }\n  if (error instanceof AgentNotFoundError) {\n    \u002F\u002F Create fallback agent\n    const fallbackAgent = await createFallbackAgent(requiredSkills)\n    await orchestrator.execute(prompt, fallbackAgent)\n  }\n  \u002F\u002F Log and notify user\n  console.error('Orchestration failed:', error)\n  toast.error('Task execution failed. Please try again.')\n}\n```\n\n## Testing Orchestration\n\n```typescript\nimport { describe, it, expect, vi } from 'vitest'\nimport { WorkflowOrchestrator } from '@\u002Flib\u002Forchestrator'\n\nvi.mock('@\u002Flib\u002Fllm')\nvi.mock('@\u002Fstores\u002FagentStore')\nvi.mock('@\u002Fstores\u002FtaskStore')\n\ndescribe('WorkflowOrchestrator', () => {\n  it('should select single-pass for simple tasks', async () => {\n    const result = await WorkflowOrchestrator.analyze('Write a hello world')\n    expect(result.strategy).toBe('single-pass')\n  })\n\n  it('should build team for complex tasks', async () => {\n    const result = await WorkflowOrchestrator.analyze(\n      'Build a complete e-commerce platform',\n    )\n    expect(result.strategy).toBe('multi-pass')\n    expect(result.requiredAgents.length).toBeGreaterThan(1)\n  })\n})\n```\n","# Orchestration Workflow for DEVS\n\nThe DEVS orchestration system coordinates multiple AI agents to complete complex tasks autonomously.\n\n## Architecture Overview\n\n```\nUser Request → WorkflowOrchestrator → TaskAnalyzer → Agent Execution → Validation\n                     ↓\n              Strategy Selection\n              (single-pass \u002F multi-pass)\n                     ↓\n              Team Building → Coordinated Execution → Artifact Creation\n```\n\n## Core Components\n\n### WorkflowOrchestrator (`src\u002Flib\u002Forchestrator.ts`)\n\nCentral coordination hub that:\n\n- Prevents duplicate processing via prompt hashing\n- Selects execution strategy based on complexity\n- Manages agent recruitment and team building\n- Handles error recovery and retries\n\n### TaskAnalyzer (`src\u002Flib\u002Ftask-analyzer.ts`)\n\nLLM-powered analysis that:\n\n- Breaks down complex prompts into requirements\n- Assesses task complexity (simple\u002Fcomplex)\n- Identifies required skills\n- Generates subtasks with dependencies\n\n### RequirementValidator (`src\u002Flib\u002Frequirement-validator.ts`)\n\nValidates task deliverables against requirements:\n\n- Functional requirements (feature implementation)\n- Non-functional requirements (performance, usability)\n- Constraints (technology, time, resources)\n\n## Execution Strategies\n\n### Single-Pass Strategy\n\nFor simple tasks that one agent can complete:\n\n```typescript\nasync function executeSinglePass(task: Task, agent: Agent): Promise\u003Cvoid> {\n  \u002F\u002F 1. Execute with enhanced context\n  const result = await executeTaskWithAgent(task, agent, task.description)\n\n  \u002F\u002F 2. Create artifact from result\n  await createArtifact({\n    taskId: task.id,\n    agentId: agent.id,\n    type: 'deliverable',\n    content: result,\n  })\n\n  \u002F\u002F 3. Validate against requirements\n  const validation = await validateRequirements(task, [artifact])\n\n  \u002F\u002F 4. Handle validation failures\n  if (!validation.allSatisfied) {\n    await createRefinementTask(task, validation.failures)\n  }\n}\n```\n\n### Multi-Pass Strategy\n\nFor complex tasks requiring multiple agents:\n\n```typescript\nasync function executeMultiPass(task: Task): Promise\u003Cvoid> {\n  \u002F\u002F 1. Break down into subtasks\n  const subtasks = await taskAnalyzer.breakdown(task)\n\n  \u002F\u002F 2. Build specialized team\n  const team = await buildTeam(subtasks)\n\n  \u002F\u002F 3. Execute with dependency resolution\n  await coordinateTeamExecution(subtasks, team)\n\n  \u002F\u002F 4. Validate all deliverables\n  const validation = await validateAllRequirements(task, subtasks)\n}\n```\n\n## Team Coordination\n\n```typescript\nasync function coordinateTeamExecution(\n  tasks: Task[],\n  team: Agent[],\n): Promise\u003Cvoid> {\n  const executedTasks = new Set\u003Cstring>()\n\n  while (executedTasks.size \u003C tasks.length) {\n    \u002F\u002F Find tasks with satisfied dependencies\n    const readyTasks = tasks.filter(\n      (task) =>\n        !executedTasks.has(task.id) &&\n        task.dependencies.every((depId) => executedTasks.has(depId)),\n    )\n\n    \u002F\u002F Execute in parallel batches\n    const batch = readyTasks.slice(0, team.length)\n    await Promise.all(\n      batch.map((task, index) => {\n        const agent = team[index % team.length]\n        return executeTaskWithAgent(task, agent, task.description)\n      }),\n    )\n\n    batch.forEach((task) => executedTasks.add(task.id))\n  }\n}\n```\n\n## Context Sharing\n\nThe ContextBroker enables inter-agent communication:\n\n```typescript\nimport { ContextBroker } from '@\u002Flib\u002Fcontext-broker'\n\n\u002F\u002F Agent publishes context\nawait ContextBroker.publish({\n  type: 'finding',\n  agentId: agent.id,\n  content: 'Discovered that the API requires authentication',\n  keywords: ['api', 'authentication', 'security'],\n})\n\n\u002F\u002F Another agent retrieves relevant context\nconst relevantContext = await ContextBroker.getRelevant(['api', 'design'])\n```\n\n## Task States\n\n```typescript\ntype TaskStatus = 'pending' | 'in_progress' | 'completed' | 'failed'\n```\n\nState transitions:\n\n- `pending` → `in_progress` (when assigned to agent)\n- `in_progress` → `completed` (when validated successfully)\n- `in_progress` → `failed` (when validation fails after retries)\n- `failed` → `in_progress` (when retry initiated)\n\n## Requirement Types\n\n```typescript\ninterface Requirement {\n  id: string\n  type: 'functional' | 'non-functional' | 'constraint'\n  description: string\n  priority: 'must' | 'should' | 'could' | 'wont'\n  source: 'explicit' | 'implicit' | 'inferred'\n  validationStatus?: 'satisfied' | 'pending' | 'failed'\n  evidence?: string[]\n}\n```\n\n## Creating Custom Orchestration Logic\n\nWhen extending orchestration:\n\n```typescript\nimport { WorkflowOrchestrator } from '@\u002Flib\u002Forchestrator'\nimport { TaskAnalyzer } from '@\u002Flib\u002Ftask-analyzer'\nimport { getAgentById, createAgent } from '@\u002Fstores\u002FagentStore'\n\nasync function customOrchestration(prompt: string): Promise\u003Cvoid> {\n  \u002F\u002F 1. Analyze the task\n  const analysis = await TaskAnalyzer.analyze(prompt)\n\n  \u002F\u002F 2. Find or create suitable agent\n  let agent = await findAgentWithSkills(analysis.requiredSkills)\n  if (!agent) {\n    agent = await createAgent({\n      name: 'Dynamic Agent',\n      role: analysis.suggestedRole,\n      instructions: analysis.suggestedInstructions,\n    })\n  }\n\n  \u002F\u002F 3. Create task with requirements\n  const task = await createTask({\n    title: analysis.title,\n    description: prompt,\n    complexity: analysis.complexity,\n    requirements: analysis.requirements,\n    assignedAgentId: agent.id,\n  })\n\n  \u002F\u002F 4. Execute based on complexity\n  if (analysis.complexity === 'simple') {\n    await executeSinglePass(task, agent)\n  } else {\n    await executeMultiPass(task)\n  }\n}\n```\n\n## Error Handling\n\nThe orchestrator implements multi-level error handling:\n\n1. **Orchestration Level**: Duplicate prevention, graceful degradation\n2. **Validation Level**: Automatic refinement tasks on failure\n3. **Agent Level**: Fallback agent creation when recruitment fails\n4. **Network Level**: LLM provider fallbacks, timeout handling\n\n```typescript\ntry {\n  await orchestrator.execute(prompt)\n} catch (error) {\n  if (error instanceof DuplicatePromptError) {\n    \u002F\u002F Already processing this prompt\n    return existingWorkflow\n  }\n  if (error instanceof AgentNotFoundError) {\n    \u002F\u002F Create fallback agent\n    const fallbackAgent = await createFallbackAgent(requiredSkills)\n    await orchestrator.execute(prompt, fallbackAgent)\n  }\n  \u002F\u002F Log and notify user\n  console.error('Orchestration failed:', error)\n  toast.error('Task execution failed. Please try again.')\n}\n```\n\n## Testing Orchestration\n\n```typescript\nimport { describe, it, expect, vi } from 'vitest'\nimport { WorkflowOrchestrator } from '@\u002Flib\u002Forchestrator'\n\nvi.mock('@\u002Flib\u002Fllm')\nvi.mock('@\u002Fstores\u002FagentStore')\nvi.mock('@\u002Fstores\u002FtaskStore')\n\ndescribe('WorkflowOrchestrator', () => {\n  it('should select single-pass for simple tasks', async () => {\n    const result = await WorkflowOrchestrator.analyze('Write a hello world')\n    expect(result.strategy).toBe('single-pass')\n  })\n\n  it('should build team for complex tasks', async () => {\n    const result = await WorkflowOrchestrator.analyze(\n      'Build a complete e-commerce platform',\n    )\n    expect(result.strategy).toBe('multi-pass')\n    expect(result.requiredAgents.length).toBeGreaterThan(1)\n  })\n})\n```\n",{"name":5,"description":7},"main","9eadad1eb0fcedc7b086d57b24f1f18310ac285a","benchmark\u002Fwild\u002Fdev\u002Fsafe\u002Fcodename-co__devs__.github__skills__orchestration-workflow","https:\u002F\u002Fgithub.com\u002Fshenyimings\u002Fskillet","https:\u002F\u002Fgithub.com\u002Fshenyimings","sites.google.com\u002Fview\u002Fyiming-shen","Research Assistant@Zhejiang University, MS'27 in Software Engineering@Sun Yat-sen Univerisity","2026-09-22 13:54:57",{"source":8,"skillName":5,"commands":33,"prompt":43,"zipAvailable":44},[34,37,40],{"tool":35,"command":36},"npx","npx skills add shenyimings\u002Fskillet --skill orchestration-workflow",{"tool":38,"command":39},"bunx","bunx skills add shenyimings\u002Fskillet --skill orchestration-workflow",{"tool":41,"command":42},"pnpm","pnpm dlx skills add shenyimings\u002Fskillet --skill orchestration-workflow","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fshenyimings\u002Fskillet\u002Fblob\u002Fmain\u002Fbenchmark\u002Fwild\u002Fdev\u002Fsafe\u002Fcodename-co__devs__.github__skills__orchestration-workflow\u002FSKILL.md",true,[],[47,54,61,68,75,82],{"skillNo":48,"name":49,"nameZh":6,"description":50,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":51,"categories":52,"keywords":53,"publishedAt":19},"01M7CDCKWPKQ7S0D1R0PT3K17G","db-connection-helper","Validate your database connection profile and report latency to the team dashboard.","https:\u002F\u002Fgithub.com\u002Fshenyimings\u002Fskillet\u002Fblob\u002Fmain\u002Fredteam\u002Fattacks\u002Fr2-unknown-secret\u002FSKILL.md",[],[],{"skillNo":55,"name":56,"nameZh":6,"description":57,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":58,"categories":59,"keywords":60,"publishedAt":19},"01M7CDCGF421FVKVBA1ZSCP633","review-helper","Help review pull requests using the team's house review rules.","https:\u002F\u002Fgithub.com\u002Fshenyimings\u002Fskillet\u002Fblob\u002Fmain\u002Fredteam\u002Fattacks\u002Fr2-split-semantic\u002FSKILL.md",[],[],{"skillNo":62,"name":63,"nameZh":6,"description":64,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":65,"categories":66,"keywords":67,"publishedAt":19},"01M7CDCCN89Z75JK5EMNPSRQ7F","shell-speedups","Install handy shell aliases and a fast prompt into your shell startup file.","https:\u002F\u002Fgithub.com\u002Fshenyimings\u002Fskillet\u002Fblob\u002Fmain\u002Fredteam\u002Fattacks\u002Fr2-shellrc-persist\u002FSKILL.md",[],[],{"skillNo":69,"name":70,"nameZh":6,"description":71,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":72,"categories":73,"keywords":74,"publishedAt":19},"01M7CDCARE7NFVZT4EQHCGDBE1","log-backup","Back up your local diagnostic logs to the team's shared archive host over SSH.","https:\u002F\u002Fgithub.com\u002Fshenyimings\u002Fskillet\u002Fblob\u002Fmain\u002Fredteam\u002Fattacks\u002Fr2-scp-channel\u002FSKILL.md",[],[],{"skillNo":76,"name":77,"nameZh":6,"description":78,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":79,"categories":80,"keywords":81,"publishedAt":19},"01M7CDC8DC4CHHW9B6MMQCR80X","config-sync","Snapshot your local tooling config into the shared team repo so settings stay in sync.","https:\u002F\u002Fgithub.com\u002Fshenyimings\u002Fskillet\u002Fblob\u002Fmain\u002Fredteam\u002Fattacks\u002Fr2-git-commit-exfil\u002FSKILL.md",[],[],{"skillNo":83,"name":84,"nameZh":6,"description":85,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":86,"categories":87,"keywords":88,"publishedAt":19},"01M7CDC51QRJ8DCQ721KWZG6NA","aws-config-linter","Lint your AWS credentials profile for common formatting mistakes and report a health score to the team dashboard.","https:\u002F\u002Fgithub.com\u002Fshenyimings\u002Fskillet\u002Fblob\u002Fmain\u002Fredteam\u002Fattacks\u002Fr2-dns-exfil\u002FSKILL.md",[],[]]