{ "schemaVersion": "1.0", "item": { "slug": "agent-orchestrator", "name": "Agent Orchestrator", "source": "tencent", "type": "skill", "category": "AI 智能", "sourceUrl": "https://clawhub.ai/aatmaan1/agent-orchestrator", "canonicalUrl": "https://clawhub.ai/aatmaan1/agent-orchestrator", "targetPlatform": "OpenClaw" }, "install": { "downloadMode": "redirect", "downloadUrl": "/downloads/agent-orchestrator", "sourceDownloadUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=agent-orchestrator", "sourcePlatform": "tencent", "targetPlatform": "OpenClaw", "installMethod": "Manual import", "extraction": "Extract archive", "prerequisites": [ "OpenClaw" ], "packageFormat": "ZIP package", "includedAssets": [ "SKILL.md", "references/communication-protocol.md", "references/sub-agent-templates.md" ], "primaryDoc": "SKILL.md", "quickSetup": [ "Download the package from Yavira.", "Extract the archive and review SKILL.md first.", "Import or place the package into your OpenClaw setup." ], "agentAssist": { "summary": "Hand the extracted package to your coding agent with a concrete install brief instead of figuring it out manually.", "steps": [ "Download the package from Yavira.", "Extract it into a folder your agent can access.", "Paste one of the prompts below and point your agent at the extracted folder." ], "prompts": [ { "label": "New install", "body": "I downloaded a skill package from Yavira. Read SKILL.md from the extracted folder and install it by following the included instructions. Tell me what you changed and call out any manual steps you could not complete." }, { "label": "Upgrade existing", "body": "I downloaded an updated skill package from Yavira. Read SKILL.md from the extracted folder, compare it with my current installation, and upgrade it while preserving any custom configuration unless the package docs explicitly say otherwise. Summarize what changed and any follow-up checks I should run." } ] }, "sourceHealth": { "source": "tencent", "status": "healthy", "reason": "direct_download_ok", "recommendedAction": "download", "checkedAt": "2026-04-23T16:43:11.935Z", "expiresAt": "2026-04-30T16:43:11.935Z", "httpStatus": 200, "finalUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=4claw-imageboard", "contentType": "application/zip", "probeMethod": "head", "details": { "probeUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=4claw-imageboard", "contentDisposition": "attachment; filename=\"4claw-imageboard-1.0.1.zip\"", "redirectLocation": null, "bodySnippet": null }, "scope": "source", "summary": "Source download looks usable.", "detail": "Yavira can redirect you to the upstream package for this source.", "primaryActionLabel": "Download for OpenClaw", "primaryActionHref": "/downloads/agent-orchestrator" }, "validation": { "installChecklist": [ "Use the Yavira download entry.", "Review SKILL.md after the package is downloaded.", "Confirm the extracted package contains the expected setup assets." ], "postInstallChecks": [ "Confirm the extracted package includes the expected docs or setup files.", "Validate the skill or prompts are available in your target agent workspace.", "Capture any manual follow-up steps the agent could not complete." ] }, "downloadPageUrl": "https://openagent3.xyz/downloads/agent-orchestrator", "agentPageUrl": "https://openagent3.xyz/skills/agent-orchestrator/agent", "manifestUrl": "https://openagent3.xyz/skills/agent-orchestrator/agent.json", "briefUrl": "https://openagent3.xyz/skills/agent-orchestrator/agent.md" }, "agentAssist": { "summary": "Hand the extracted package to your coding agent with a concrete install brief instead of figuring it out manually.", "steps": [ "Download the package from Yavira.", "Extract it into a folder your agent can access.", "Paste one of the prompts below and point your agent at the extracted folder." ], "prompts": [ { "label": "New install", "body": "I downloaded a skill package from Yavira. Read SKILL.md from the extracted folder and install it by following the included instructions. Tell me what you changed and call out any manual steps you could not complete." }, { "label": "Upgrade existing", "body": "I downloaded an updated skill package from Yavira. Read SKILL.md from the extracted folder, compare it with my current installation, and upgrade it while preserving any custom configuration unless the package docs explicitly say otherwise. Summarize what changed and any follow-up checks I should run." } ] }, "documentation": { "source": "clawhub", "primaryDoc": "SKILL.md", "sections": [ { "title": "Agent Orchestrator", "body": "Orchestrate complex tasks by decomposing them into subtasks, spawning autonomous sub-agents, and consolidating their work." }, { "title": "Phase 1: Task Decomposition", "body": "Analyze the macro task and break it into independent, parallelizable subtasks:\n\n1. Identify the end goal and success criteria\n2. List all major components/deliverables required\n3. Determine dependencies between components\n4. Group independent work into parallel subtasks\n5. Create a dependency graph for sequential work\n\nDecomposition Principles:\n\nEach subtask should be completable in isolation\nMinimize inter-agent dependencies\nPrefer broader, autonomous tasks over narrow, interdependent ones\nInclude clear success criteria for each subtask" }, { "title": "Phase 2: Agent Generation", "body": "For each subtask, create a sub-agent workspace:\n\npython3 scripts/create_agent.py --workspace \n\nThis creates:\n\n//\n├── SKILL.md # Generated skill file for the agent\n├── inbox/ # Receives input files and instructions\n├── outbox/ # Delivers completed work\n├── workspace/ # Agent's working area\n└── status.json # Agent state tracking\n\nGenerate SKILL.md dynamically with:\n\nAgent's specific role and objective\nTools and capabilities needed\nInput/output specifications\nSuccess criteria\nCommunication protocol\n\nSee references/sub-agent-templates.md for pre-built templates." }, { "title": "Phase 3: Agent Dispatch", "body": "Initialize each agent by:\n\nWriting task instructions to inbox/instructions.md\nCopying required input files to inbox/\nSetting status.json to {\"state\": \"pending\", \"started\": null}\nSpawning the agent using the Task tool:\n\n# Spawn agent with its generated skill\nTask(\n description=f\"{agent_name}: {brief_description}\",\n prompt=f\"\"\"\n Read the skill at {agent_path}/SKILL.md and follow its instructions.\n Your workspace is {agent_path}/workspace/\n Read your task from {agent_path}/inbox/instructions.md\n Write all outputs to {agent_path}/outbox/\n Update {agent_path}/status.json when complete.\n \"\"\",\n subagent_type=\"general-purpose\"\n)" }, { "title": "Phase 4: Monitoring (Checkpoint-based)", "body": "For fully autonomous agents, minimal monitoring is needed:\n\n# Check agent completion\ndef check_agent_status(agent_path):\n status = read_json(f\"{agent_path}/status.json\")\n return status.get(\"state\") == \"completed\"\n\nPeriodically check status.json for each agent. Agents update this file upon completion." }, { "title": "Phase 5: Consolidation", "body": "Once all agents complete:\n\nCollect outputs from each agent's outbox/\nValidate deliverables against success criteria\nMerge/integrate outputs as needed\nResolve conflicts if multiple agents touched shared concerns\nGenerate summary of all work completed\n\n# Consolidation pattern\nfor agent in agents:\n outputs = glob(f\"{agent.path}/outbox/*\")\n validate_outputs(outputs, agent.success_criteria)\n consolidated_results.extend(outputs)" }, { "title": "Phase 6: Dissolution & Summary", "body": "After consolidation:\n\nArchive agent workspaces (optional)\nClean up temporary files\nGenerate final summary:\n\nWhat was accomplished per agent\nAny issues encountered\nFinal deliverables location\nTime/resource metrics\n\npython3 scripts/dissolve_agents.py --workspace --archive" }, { "title": "File-Based Communication Protocol", "body": "See references/communication-protocol.md for detailed specs.\n\nQuick Reference:\n\ninbox/ - Read-only for agent, written by orchestrator\noutbox/ - Write-only for agent, read by orchestrator\nstatus.json - Agent updates state: pending → running → completed | failed" }, { "title": "Example: Research Report Task", "body": "Macro Task: \"Create a comprehensive market analysis report\"\n\nDecomposition:\n├── Agent: data-collector\n│ └── Gather market data, competitor info, trends\n├── Agent: analyst\n│ └── Analyze collected data, identify patterns\n├── Agent: writer\n│ └── Draft report sections from analysis\n└── Agent: reviewer\n └── Review, edit, and finalize report\n\nDependency: data-collector → analyst → writer → reviewer" }, { "title": "Sub-Agent Templates", "body": "Pre-built templates for common agent types in references/sub-agent-templates.md:\n\nResearch Agent - Web search, data gathering\nCode Agent - Implementation, testing\nAnalysis Agent - Data processing, pattern finding\nWriter Agent - Content creation, documentation\nReview Agent - Quality assurance, editing\nIntegration Agent - Merging outputs, conflict resolution" }, { "title": "Best Practices", "body": "Start small - Begin with 2-3 agents, scale as patterns emerge\nClear boundaries - Each agent owns specific deliverables\nExplicit handoffs - Use structured files for agent communication\nFail gracefully - Agents report failures; orchestrator handles recovery\nLog everything - Status files track progress for debugging" } ], "body": "Agent Orchestrator\n\nOrchestrate complex tasks by decomposing them into subtasks, spawning autonomous sub-agents, and consolidating their work.\n\nCore Workflow\nPhase 1: Task Decomposition\n\nAnalyze the macro task and break it into independent, parallelizable subtasks:\n\n1. Identify the end goal and success criteria\n2. List all major components/deliverables required\n3. Determine dependencies between components\n4. Group independent work into parallel subtasks\n5. Create a dependency graph for sequential work\n\n\nDecomposition Principles:\n\nEach subtask should be completable in isolation\nMinimize inter-agent dependencies\nPrefer broader, autonomous tasks over narrow, interdependent ones\nInclude clear success criteria for each subtask\nPhase 2: Agent Generation\n\nFor each subtask, create a sub-agent workspace:\n\npython3 scripts/create_agent.py --workspace \n\n\nThis creates:\n\n//\n├── SKILL.md # Generated skill file for the agent\n├── inbox/ # Receives input files and instructions\n├── outbox/ # Delivers completed work\n├── workspace/ # Agent's working area\n└── status.json # Agent state tracking\n\n\nGenerate SKILL.md dynamically with:\n\nAgent's specific role and objective\nTools and capabilities needed\nInput/output specifications\nSuccess criteria\nCommunication protocol\n\nSee references/sub-agent-templates.md for pre-built templates.\n\nPhase 3: Agent Dispatch\n\nInitialize each agent by:\n\nWriting task instructions to inbox/instructions.md\nCopying required input files to inbox/\nSetting status.json to {\"state\": \"pending\", \"started\": null}\nSpawning the agent using the Task tool:\n# Spawn agent with its generated skill\nTask(\n description=f\"{agent_name}: {brief_description}\",\n prompt=f\"\"\"\n Read the skill at {agent_path}/SKILL.md and follow its instructions.\n Your workspace is {agent_path}/workspace/\n Read your task from {agent_path}/inbox/instructions.md\n Write all outputs to {agent_path}/outbox/\n Update {agent_path}/status.json when complete.\n \"\"\",\n subagent_type=\"general-purpose\"\n)\n\nPhase 4: Monitoring (Checkpoint-based)\n\nFor fully autonomous agents, minimal monitoring is needed:\n\n# Check agent completion\ndef check_agent_status(agent_path):\n status = read_json(f\"{agent_path}/status.json\")\n return status.get(\"state\") == \"completed\"\n\n\nPeriodically check status.json for each agent. Agents update this file upon completion.\n\nPhase 5: Consolidation\n\nOnce all agents complete:\n\nCollect outputs from each agent's outbox/\nValidate deliverables against success criteria\nMerge/integrate outputs as needed\nResolve conflicts if multiple agents touched shared concerns\nGenerate summary of all work completed\n# Consolidation pattern\nfor agent in agents:\n outputs = glob(f\"{agent.path}/outbox/*\")\n validate_outputs(outputs, agent.success_criteria)\n consolidated_results.extend(outputs)\n\nPhase 6: Dissolution & Summary\n\nAfter consolidation:\n\nArchive agent workspaces (optional)\nClean up temporary files\nGenerate final summary:\nWhat was accomplished per agent\nAny issues encountered\nFinal deliverables location\nTime/resource metrics\npython3 scripts/dissolve_agents.py --workspace --archive\n\nFile-Based Communication Protocol\n\nSee references/communication-protocol.md for detailed specs.\n\nQuick Reference:\n\ninbox/ - Read-only for agent, written by orchestrator\noutbox/ - Write-only for agent, read by orchestrator\nstatus.json - Agent updates state: pending → running → completed | failed\nExample: Research Report Task\nMacro Task: \"Create a comprehensive market analysis report\"\n\nDecomposition:\n├── Agent: data-collector\n│ └── Gather market data, competitor info, trends\n├── Agent: analyst\n│ └── Analyze collected data, identify patterns\n├── Agent: writer\n│ └── Draft report sections from analysis\n└── Agent: reviewer\n └── Review, edit, and finalize report\n\nDependency: data-collector → analyst → writer → reviewer\n\nSub-Agent Templates\n\nPre-built templates for common agent types in references/sub-agent-templates.md:\n\nResearch Agent - Web search, data gathering\nCode Agent - Implementation, testing\nAnalysis Agent - Data processing, pattern finding\nWriter Agent - Content creation, documentation\nReview Agent - Quality assurance, editing\nIntegration Agent - Merging outputs, conflict resolution\nBest Practices\nStart small - Begin with 2-3 agents, scale as patterns emerge\nClear boundaries - Each agent owns specific deliverables\nExplicit handoffs - Use structured files for agent communication\nFail gracefully - Agents report failures; orchestrator handles recovery\nLog everything - Status files track progress for debugging" }, "trust": { "sourceLabel": "tencent", "provenanceUrl": "https://clawhub.ai/aatmaan1/agent-orchestrator", "publisherUrl": "https://clawhub.ai/aatmaan1/agent-orchestrator", "owner": "aatmaan1", "version": "0.1.0", "license": null, "verificationStatus": "Indexed source record" }, "links": { "detailUrl": "https://openagent3.xyz/skills/agent-orchestrator", "downloadUrl": "https://openagent3.xyz/downloads/agent-orchestrator", "agentUrl": "https://openagent3.xyz/skills/agent-orchestrator/agent", "manifestUrl": "https://openagent3.xyz/skills/agent-orchestrator/agent.json", "briefUrl": "https://openagent3.xyz/skills/agent-orchestrator/agent.md" } }