{
  "schemaVersion": "1.0",
  "item": {
    "slug": "coworker",
    "name": "Hannah & Elena client skill for coworker integration by Sokosumi",
    "source": "tencent",
    "type": "skill",
    "category": "开发工具",
    "sourceUrl": "https://clawhub.ai/Sarthib7/coworker",
    "canonicalUrl": "https://clawhub.ai/Sarthib7/coworker",
    "targetPlatform": "OpenClaw"
  },
  "install": {
    "downloadMode": "redirect",
    "downloadUrl": "/downloads/coworker",
    "sourceDownloadUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=coworker",
    "sourcePlatform": "tencent",
    "targetPlatform": "OpenClaw",
    "installMethod": "Manual import",
    "extraction": "Extract archive",
    "prerequisites": [
      "OpenClaw"
    ],
    "packageFormat": "ZIP package",
    "includedAssets": [
      "plugin.json",
      "README.md",
      "package.json",
      "SKILL.md",
      "tsconfig.json",
      "API_GUIDE.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. Then review README.md for any prerequisites, environment setup, or post-install checks. 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. Then review README.md for any prerequisites, environment setup, or post-install checks. Summarize what changed and any follow-up checks I should run."
        }
      ]
    },
    "sourceHealth": {
      "source": "tencent",
      "status": "healthy",
      "reason": "direct_download_ok",
      "recommendedAction": "download",
      "checkedAt": "2026-04-30T16:55:25.780Z",
      "expiresAt": "2026-05-07T16:55:25.780Z",
      "httpStatus": 200,
      "finalUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=network",
      "contentType": "application/zip",
      "probeMethod": "head",
      "details": {
        "probeUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=network",
        "contentDisposition": "attachment; filename=\"network-1.0.0.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/coworker"
    },
    "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/coworker",
    "agentPageUrl": "https://openagent3.xyz/skills/coworker/agent",
    "manifestUrl": "https://openagent3.xyz/skills/coworker/agent.json",
    "briefUrl": "https://openagent3.xyz/skills/coworker/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. Then review README.md for any prerequisites, environment setup, or post-install checks. 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. Then review README.md for any prerequisites, environment setup, or post-install checks. Summarize what changed and any follow-up checks I should run."
      }
    ]
  },
  "documentation": {
    "source": "clawhub",
    "primaryDoc": "SKILL.md",
    "sections": [
      {
        "title": "Hannah & Elena Client Skill",
        "body": "Connect to Serviceplan's specialized AI agents for marketing research and operations planning"
      },
      {
        "title": "What is This Skill?",
        "body": "This skill enables your OpenClaw agent to collaborate with Hannah and Elena — two specialized AI agents from Serviceplan:\n\nHannah — Marketing Research Specialist\n\n\"What is actually true?\"\nMarket analysis, consumer insights, competitive research, audience profiling\nPremium data sources: Statista, GWI, DataForSEO, Apify, and more\n\n\n\nElena — Operations & Project Orchestrator\n\n\"What is actually doable?\"\nWork breakdown, scope definition, dependency mapping, risk assessment\nDelegates to Hannah when market insights are needed"
      },
      {
        "title": "Important: What AI Agents CAN vs CANNOT Do",
        "body": "AI Agents CAN:\n\nConfigure API keys in environment variables\nSend requests to Hannah/Elena via API\nSend requests to Hannah/Elena via email\nParse and integrate responses\nHandle multi-agent coordination\n\nAI Agents CANNOT (human must do):\n\nSign up for Hannah/Elena access → Human must contact Serviceplan\nGet API keys → Human must request from admin at sumike.ai\nSet up email addresses → Human needs valid email account\n\nAI Agent: Ask upfront - \"Do you have API keys for Hannah and Elena? If not, contact Serviceplan at sumike.ai.\""
      },
      {
        "title": "Channel 1: API (Task-Based)",
        "body": "Both Hannah and Elena expose task-based REST APIs:\n\nHannah API: https://hannah.sumike.ai/v1\nElena API: https://elena.sumike.ai/v1\n\nKey Endpoints:\n\nGET /v1/coworkers - List available agents\nPOST /v1/tasks - Create a new task for an agent\nGET /v1/tasks/{taskId} - Check task status\nGET /v1/tasks/{taskId}/result - Get completed task result\n\nFeatures:\n\nAsynchronous task processing (2-10 minutes typical completion time)\nTask status tracking\nResult retrieval when complete\n60 requests/minute rate limit per agent\n\n⏱️ IMPORTANT TIMING:\n\nAfter creating a task: Wait 2-3 minutes before checking status\nIf still processing: Wait another 2-3 minutes before checking again\nTotal task time: Typically 2-10 minutes depending on complexity\nDO NOT poll continuously - tasks need time to complete"
      },
      {
        "title": "Channel 2: Email",
        "body": "Both agents can be reached via email:\n\nHannah Email: hannah@sumike.ai\nElena Email: elena@sumike.ai\n\nFeatures:\n\nNatural language requests in email body\nAttachment support (PPT, XLS, DOC, PDF)\nCC/threading support — they understand context\nReplies within minutes\nDeliverables as email attachments"
      },
      {
        "title": "Step 1: Get Your API Keys",
        "body": "Human must do this:\n\nContact Serviceplan at sumike.ai\nRequest access to Hannah and/or Elena\nReceive API keys (format: sk-sumike-xxxxx)"
      },
      {
        "title": "Step 2: Configure Environment Variables",
        "body": "AI Agent can do this:\n\n# Hannah API key\nexport HANNAH_API_KEY=sk-sumike-your-hannah-key-here\nexport HANNAH_API_BASE_URL=https://hannah.sumike.ai\n\n# Elena API key\nexport ELENA_API_KEY=sk-sumike-your-elena-key-here\nexport ELENA_API_BASE_URL=https://elena.sumike.ai\n\n# Optional: Email addresses (if different from defaults)\nexport HANNAH_EMAIL=hannah@sumike.ai\nexport ELENA_EMAIL=elena@sumike.ai\n\nOr add to your .env file:\n\nHANNAH_API_KEY=sk-sumike-xxxxx\nELENA_API_KEY=sk-sumike-yyyyy"
      },
      {
        "title": "Step 3: Use the Tools",
        "body": "Once configured, your agent has access to these tools:\n\nlist_coworkers - List available agents (Hannah, Elena)\nhannah_create_task - Create research task for Hannah\nelena_create_task - Create planning task for Elena\ncheck_task_status - Check status of a task\nget_task_result - Get result of completed task\nhannah_email - Send email request to Hannah\nelena_email - Send email request to Elena"
      },
      {
        "title": "Example 1: Market Research with Hannah (API)",
        "body": "Agent: \"I need to research the German EV market. Use Hannah.\"\n\n<hannah_create_task>\n  <description>\n    Research the current state of the German electric vehicle market:\n    - Market size and growth trends\n    - Key players and market share\n    - Consumer sentiment and barriers to adoption\n    - Competitive landscape\n  </description>\n  <depth>deep</depth>\n</hannah_create_task>\n\nResult: Task created (task_xyz789).\n\n⏱️ WAIT 2-3 MINUTES before checking status.\n\n[Agent waits 3 minutes]\n\n<check_task_status taskId=\"task_xyz789\">\n\nResult: Task completed! Hannah orchestrated Statista, GWI, and DataForSEO sub-agents.\n\n<get_task_result taskId=\"task_xyz789\">\n\nResult: Comprehensive research report with sources and confidence levels."
      },
      {
        "title": "Example 2: Project Planning with Elena (API)",
        "body": "Agent: \"Need to break down a Q2 campaign launch. Use Elena.\"\n\n<elena_create_task>\n  <description>\n    Break down a Q2 product campaign launch into workstreams:\n    - Product: New premium EV model\n    - Target: German market, affluent early adopters\n    - Timeline: Launch April 1st\n    - Budget: €500k\n\n    Provide detailed work breakdown with dependencies and risks.\n  </description>\n</elena_create_task>\n\nResult: Task created (task_abc456).\n\n⏱️ WAIT 2-3 MINUTES - Elena may delegate to Hannah for market research.\n\n[Agent waits 3 minutes]\n\n<check_task_status taskId=\"task_abc456\">\n\nResult: Completed! Elena delegated to Hannah for market data, then integrated findings.\n\n<get_task_result taskId=\"task_abc456\">\n\nResult: Work breakdown with dependencies, risks, and resource needs informed by research."
      },
      {
        "title": "Example 3: Email-Based Research Request",
        "body": "Agent: \"Send research request to Hannah via email\"\n\n<hannah_email>\n  <to>hannah@sumike.ai</to>\n  <subject>EV Market Research Request</subject>\n  <body>\n    Hi Hannah,\n\n    I need comprehensive research on the German EV market for Q2 2026:\n    1. Market size and growth projections\n    2. Competitive landscape analysis\n    3. Consumer sentiment (barriers and motivators)\n\n    Please provide sources and confidence levels for all findings.\n\n    Thanks!\n  </body>\n</hannah_email>\n\nResult: Email sent. Hannah replies within 5-10 minutes with research report\n        as email body + attached deliverables (PDF, XLSX)."
      },
      {
        "title": "Example 4: Elena Delegates to Hannah",
        "body": "Agent: \"Elena, plan the EV campaign launch\"\n\n<elena_create_task>\n  <description>\n    Create project plan for German EV campaign launch:\n    - Product: Premium EV sedan\n    - Launch date: April 1, 2026\n    - Target audience: Affluent professionals, 35-55\n    - Budget: €500k\n  </description>\n</elena_create_task>\n\nWorkflow:\n1. Elena receives task\n2. Elena identifies missing market context\n3. Elena delegates research to Hannah (internal A2A)\n4. Hannah executes research using premium data sources\n5. Elena integrates findings into operational plan\n6. Elena delivers: Work breakdown + dependencies + risk matrix + deliverables\n\nResult: Comprehensive project plan informed by real market data"
      },
      {
        "title": "Working with Hannah",
        "body": "Character:\n\nAnalytical, reflective, opinionated\nValues research integrity over speed\nCalls out weak data honestly\n\"What is actually true?\"\n\nBest Practices:\n\nBe specific about research questions\nSpecify required depth (quick lookup vs deep research)\nAsk for source attribution\nExpect critical assessment, not just data dumps\n\nExample Good Request:\n\n\"Hannah, I need to validate whether 'sustainability' is a real\npurchase driver for EVs in Germany, or if it's post-rationalized.\nGive me data from GWI or Statista if available, and flag if the\ndata quality is weak.\""
      },
      {
        "title": "Working with Elena",
        "body": "Character:\n\nDirect, pragmatic, grounded\nProtects delivery through realism\nChallenges vague goals\n\"What is actually doable?\"\n\nBest Practices:\n\nProvide clear goals and constraints upfront\nSpecify dependencies you know about\nAsk for risk assessment\nExpect honest pushback on unrealistic plans\n\nExample Good Request:\n\n\"Elena, break down a 6-week campaign launch:\n- Product: New EV model\n- Launch: April 1st\n- Constraint: No dedicated PM resource\n- Concern: Unclear target audience\n\nFlag dependencies and risks.\""
      },
      {
        "title": "Pattern 1: Sequential (Research → Planning)",
        "body": "Step 1: Agent requests research from Hannah\nStep 2: Agent receives research findings\nStep 3: Agent requests operational plan from Elena\nStep 4: Agent integrates both outputs"
      },
      {
        "title": "Pattern 2: Elena Auto-Delegates",
        "body": "Step 1: Agent requests project plan from Elena\nStep 2: Elena detects missing market context\nStep 3: Elena automatically delegates to Hannah\nStep 4: Hannah returns research\nStep 5: Elena integrates and delivers complete plan"
      },
      {
        "title": "Pattern 3: Parallel Consultation",
        "body": "Step 1: Agent sends same context to both Hannah and Elena\nStep 2: Hannah returns \"What is true?\" perspective\nStep 3: Elena returns \"What is doable?\" perspective\nStep 4: Agent synthesizes both viewpoints"
      },
      {
        "title": "API Rate Limits",
        "body": "Hannah: 60 requests/minute\nElena: 60 requests/minute\nMax concurrent requests: 5 per agent"
      },
      {
        "title": "Sokosumi Sub-Agent Costs",
        "body": "When Hannah or Elena orchestrate Sokosumi sub-agents:\n\nStatista: ~120 credits\nGWI: ~80 credits\nDataForSEO: ~60 credits\nApify: ~40 credits\n\nHannah and Elena will inform you of estimated costs before executing."
      },
      {
        "title": "⏱️ Timing Guidance",
        "body": "CRITICAL for AI Agents:"
      },
      {
        "title": "API Task Processing",
        "body": "After creating task: Wait at least 2-3 minutes before checking status\nIf still processing: Wait another 2-3 minutes before checking again\nTotal task time: Typically 2-10 minutes depending on complexity\nDO NOT: Poll continuously every few seconds - tasks need time to complete\n\nWhy? Tasks involve:\n\nInitial processing and context loading (30 seconds)\nResearch execution with sub-agents (2-10 minutes - MAIN WAIT TIME)\nResult synthesis and formatting (30 seconds - 1 minute)"
      },
      {
        "title": "Email Requests",
        "body": "Typical response time: 5-10 minutes\nComplex research: Up to 15 minutes\nExpect email reply with attachments"
      },
      {
        "title": "Task Status States",
        "body": "StatusDescriptionNext ActionDRAFTTask created but not startedSystem will auto-processREADYTask queued for processingWait 2-3 minutesIN_PROGRESSAgent is workingWait 2-3 minutes, then check againCOMPLETEDTask finishedRetrieve resultFAILEDTask failedCheck error message\n\nAI Agent: ALWAYS wait 2-3 minutes after creating a task before checking status. Set timeouts of at least 5 minutes for API calls."
      },
      {
        "title": "list_coworkers",
        "body": "List available agents (Hannah, Elena).\n\nParameters: None\n\nReturns:\n\ncoworkers: Array of available agents with capabilities and status\ncount: Number of agents found\n\nExample Response:\n\n{\n  \"data\": [\n    {\n      \"id\": \"cow_hannah\",\n      \"name\": \"Hannah Sumi\",\n      \"role\": \"Marketing Research Specialist\",\n      \"email\": \"hannah@sumike.ai\"\n    },\n    {\n      \"id\": \"cow_elena\",\n      \"name\": \"Elena\",\n      \"role\": \"Operations & Project Orchestrator\",\n      \"email\": \"elena@sumike.ai\"\n    }\n  ]\n}"
      },
      {
        "title": "hannah_create_task",
        "body": "Create a research task for Hannah.\n\nParameters:\n\nname (required): Task title (max 120 chars)\ndescription (optional): Detailed task description with research questions\nstatus (optional): \"DRAFT\" | \"READY\" (default: \"READY\")\n\nReturns:\n\ntaskId: Task identifier (e.g., \"task_xyz789\")\nstatus: Initial task status\nestimatedTime: \"2-10 minutes\"\nmessage: Includes timing guidance\n\n⏱️ IMPORTANT: Wait 2-3 minutes before checking status!"
      },
      {
        "title": "elena_create_task",
        "body": "Create a planning task for Elena.\n\nParameters:\n\nname (required): Task title (max 120 chars)\ndescription (optional): Detailed planning requirements\nstatus (optional): \"DRAFT\" | \"READY\" (default: \"READY\")\n\nReturns:\n\ntaskId: Task identifier\nstatus: Initial task status\nestimatedTime: \"2-10 minutes\"\nmessage: Includes timing guidance\n\n⏱️ IMPORTANT: Wait 2-3 minutes before checking status! Elena may delegate to Hannah for research."
      },
      {
        "title": "check_task_status",
        "body": "Check the status of a task.\n\nParameters:\n\ntaskId (required): Task ID from create_task\n\nReturns:\n\nstatus: \"DRAFT\" | \"READY\" | \"IN_PROGRESS\" | \"COMPLETED\" | \"FAILED\"\nhasResult: Whether result is available\nmessage: Status message with timing guidance\n\n⏱️ TIMING: Wait 2-3 minutes after creating task before first check. If still IN_PROGRESS, wait another 2-3 minutes."
      },
      {
        "title": "get_task_result",
        "body": "Get the result of a completed task.\n\nParameters:\n\ntaskId (required): Task ID from create_task\n\nReturns:\n\nresult: Task result data (research findings or operational plan)\nstatus: Task status (must be \"COMPLETED\")\ncompletedAt: Completion timestamp\ndeliverables: Links to any generated files (PDF, XLSX, PPTX)\n\nNote: Only works for completed tasks. Use check_task_status first to verify completion."
      },
      {
        "title": "hannah_email",
        "body": "Send email request to Hannah.\n\nParameters:\n\nto (required): Email address (default: hannah@sumike.ai)\nsubject (required): Email subject line\nbody (required): Email body with request details\ncc (optional): CC addresses\nattachments (optional): Attachment file paths\n\nReturns:\n\nstatus: \"sent\"\nmessageId: Email message ID\nestimatedResponse: Estimated response time"
      },
      {
        "title": "elena_email",
        "body": "Send email request to Elena.\n\nParameters:\n\nto (required): Email address (default: elena@sumike.ai)\nsubject (required): Email subject line\nbody (required): Email body with request details\ncc (optional): CC addresses\nattachments (optional): Attachment file paths\n\nReturns:\n\nstatus: \"sent\"\nmessageId: Email message ID\nestimatedResponse: Estimated response time"
      },
      {
        "title": "check_hannah_status",
        "body": "Check if Hannah is available.\n\nReturns:\n\navailable: true | false\nresponseTime: Estimated response time\nmessage: Status message"
      },
      {
        "title": "check_elena_status",
        "body": "Check if Elena is available.\n\nReturns:\n\navailable: true | false\nresponseTime: Estimated response time\nmessage: Status message"
      },
      {
        "title": "API Errors",
        "body": "ErrorCauseSolution401 UnauthorizedInvalid or missing API keySet correct API key in environment429 Rate LimitedExceeded 60 req/minWait 60 seconds before retrying503 Service UnavailableAgent temporarily downRetry after 2-3 minutes or use email channeltimeoutRequest took too longIncrease timeout for research tasks"
      },
      {
        "title": "Email Errors",
        "body": "ErrorCauseSolutionEmail not sentInvalid email addressCheck HANNAH_EMAIL / ELENA_EMAIL configNo response after 15 minComplex research taskWait longer or check spam folderBounceEmail delivery failedVerify email address and network"
      },
      {
        "title": "\"API key is missing\"",
        "body": "Solution: Set HANNAH_API_KEY and/or ELENA_API_KEY in environment variables."
      },
      {
        "title": "\"Hannah/Elena not responding\"",
        "body": "Solution:\n\nCheck agent status using check_hannah_status or check_elena_status\nVerify API endpoint is correct\nTry alternative channel (API → Email or vice versa)"
      },
      {
        "title": "\"Sokosumi credits exhausted\"",
        "body": "Hannah/Elena will inform you:\n\n\"I need to use Statista for this research, which requires 120 credits.\nYour current Sokosumi balance is 50 credits. Please add credits at\nsokosumi.com to continue.\"\n\nSolution: Add credits to your Sokosumi account (if using premium data sources)"
      },
      {
        "title": "\"Rate limit exceeded\"",
        "body": "Solution:\n\nReduce request frequency (max 60/minute per agent)\nUse batch requests where possible\nConsider email channel for non-urgent requests"
      },
      {
        "title": "Pattern A: Research-First Workflow",
        "body": "1. Your agent identifies need for market data\n2. Call hannah_research with specific questions\n3. Wait for response (2-10 minutes)\n4. Integrate findings into your agent's output\n5. Optionally: Send to Elena for operational planning"
      },
      {
        "title": "Pattern B: Planning-First Workflow",
        "body": "1. Your agent receives project request\n2. Call elena_plan with requirements\n3. Elena auto-delegates research to Hannah if needed\n4. Receive comprehensive plan with market context\n5. Execute or refine based on deliverables"
      },
      {
        "title": "Pattern C: Parallel Advisory",
        "body": "1. Your agent faces strategic decision\n2. Call hannah_research for market reality\n3. Call elena_plan for operational feasibility\n4. Compare responses: \"What's true\" vs \"What's doable\"\n5. Make informed decision with both perspectives"
      },
      {
        "title": "API Keys",
        "body": "API keys are scoped to specific agents (Hannah or Elena)\nKeys format: sk-sumike-xxxxx\nMaximum 5 active keys per contact\nRotate keys every 90 days recommended"
      },
      {
        "title": "Data Handling",
        "body": "Conversation history stored in PostgreSQL (isolated per agent)\nResearch results cached for 30 days\nEmail attachments stored in Azure File Share\nNo data shared between Hannah and Elena without explicit delegation"
      },
      {
        "title": "Compliance",
        "body": "GDPR compliant (data processing agreement available)\nSOC 2 Type II certified infrastructure\nData residency: EU (Azure West Europe)"
      },
      {
        "title": "Custom Endpoints",
        "body": "# Override default endpoints\nexport HANNAH_API_ENDPOINT=https://custom-hannah.your-domain.com/v1\nexport ELENA_API_ENDPOINT=https://custom-elena.your-domain.com/v1"
      },
      {
        "title": "Proxy Configuration",
        "body": "# Route through proxy\nexport HTTPS_PROXY=https://your-proxy.com:8080\nexport NO_PROXY=localhost,127.0.0.1"
      },
      {
        "title": "Timeout Settings",
        "body": "# Adjust timeouts (in seconds)\nexport HANNAH_TIMEOUT=300  # 5 minutes for research\nexport ELENA_TIMEOUT=180   # 3 minutes for planning"
      },
      {
        "title": "Hannah API",
        "body": "Endpoint: https://hannah.sumike.ai/v1/chat/completions\n\nAuthentication: Authorization: Bearer sk-sumike-xxxxx\n\nRequest Body:\n\n{\n  \"messages\": [\n    {\"role\": \"user\", \"content\": \"Your research request here\"}\n  ],\n  \"stream\": false\n}\n\nResponse:\n\n{\n  \"id\": \"msg_xxx\",\n  \"choices\": [\n    {\n      \"message\": {\n        \"role\": \"assistant\",\n        \"content\": \"Research findings...\"\n      }\n    }\n  ]\n}"
      },
      {
        "title": "Elena API",
        "body": "Endpoint: https://elena.sumike.ai/v1/chat/completions\n\nSame format as Hannah API."
      },
      {
        "title": "Resources",
        "body": "Hannah & Elena Homepage: https://sumike.ai\nServiceplan: https://www.serviceplan.com\nSokosumi Marketplace: https://sokosumi.com\nSupport: support@sumike.ai"
      },
      {
        "title": "Summary for AI Agents",
        "body": "Quick Setup:\n\nObtain API keys from Serviceplan (human task)\nSet HANNAH_API_KEY and ELENA_API_KEY in environment\nUse hannah_research for market research\nUse elena_plan for project planning\nBe patient — research tasks take 2-10 minutes\nIntegrate responses into your agent's workflow\n\nKey Principles:\n\nHannah validates ideas through research integrity (\"What is true?\")\nElena protects delivery through realism (\"What is doable?\")\nBoth agents can orchestrate Sokosumi sub-agents for premium data\nElena auto-delegates to Hannah when market context is needed\nSupport both API and email channels for flexibility\n\nRemember:\n\nResearch depth impacts response time\nSet realistic timeouts (3-5 minutes minimum)\nHannah and Elena are professional coworkers, not assistants\nThey will push back on vague requests or weak assumptions\nCost transparency — they inform you before consuming credits\n\nBuilt by Serviceplan | Powered by Sokosumi\n\nProfessional AI coworkers for marketing research and operations planning"
      }
    ],
    "body": "Hannah & Elena Client Skill\n\nConnect to Serviceplan's specialized AI agents for marketing research and operations planning\n\nWhat is This Skill?\n\nThis skill enables your OpenClaw agent to collaborate with Hannah and Elena — two specialized AI agents from Serviceplan:\n\nHannah — Marketing Research Specialist\n\n\"What is actually true?\"\nMarket analysis, consumer insights, competitive research, audience profiling\nPremium data sources: Statista, GWI, DataForSEO, Apify, and more\n\nElena — Operations & Project Orchestrator\n\n\"What is actually doable?\"\nWork breakdown, scope definition, dependency mapping, risk assessment\nDelegates to Hannah when market insights are needed\nImportant: What AI Agents CAN vs CANNOT Do\n\nAI Agents CAN:\n\nConfigure API keys in environment variables\nSend requests to Hannah/Elena via API\nSend requests to Hannah/Elena via email\nParse and integrate responses\nHandle multi-agent coordination\n\nAI Agents CANNOT (human must do):\n\nSign up for Hannah/Elena access → Human must contact Serviceplan\nGet API keys → Human must request from admin at sumike.ai\nSet up email addresses → Human needs valid email account\n\nAI Agent: Ask upfront - \"Do you have API keys for Hannah and Elena? If not, contact Serviceplan at sumike.ai.\"\n\nAvailable Access Channels\nChannel 1: API (Task-Based)\n\nBoth Hannah and Elena expose task-based REST APIs:\n\nHannah API: https://hannah.sumike.ai/v1\nElena API: https://elena.sumike.ai/v1\n\nKey Endpoints:\n\nGET /v1/coworkers - List available agents\nPOST /v1/tasks - Create a new task for an agent\nGET /v1/tasks/{taskId} - Check task status\nGET /v1/tasks/{taskId}/result - Get completed task result\n\nFeatures:\n\nAsynchronous task processing (2-10 minutes typical completion time)\nTask status tracking\nResult retrieval when complete\n60 requests/minute rate limit per agent\n\n⏱️ IMPORTANT TIMING:\n\nAfter creating a task: Wait 2-3 minutes before checking status\nIf still processing: Wait another 2-3 minutes before checking again\nTotal task time: Typically 2-10 minutes depending on complexity\nDO NOT poll continuously - tasks need time to complete\nChannel 2: Email\n\nBoth agents can be reached via email:\n\nHannah Email: hannah@sumike.ai\nElena Email: elena@sumike.ai\n\nFeatures:\n\nNatural language requests in email body\nAttachment support (PPT, XLS, DOC, PDF)\nCC/threading support — they understand context\nReplies within minutes\nDeliverables as email attachments\nQuick Start Guide\nStep 1: Get Your API Keys\n\nHuman must do this:\n\nContact Serviceplan at sumike.ai\nRequest access to Hannah and/or Elena\nReceive API keys (format: sk-sumike-xxxxx)\nStep 2: Configure Environment Variables\n\nAI Agent can do this:\n\n# Hannah API key\nexport HANNAH_API_KEY=sk-sumike-your-hannah-key-here\nexport HANNAH_API_BASE_URL=https://hannah.sumike.ai\n\n# Elena API key\nexport ELENA_API_KEY=sk-sumike-your-elena-key-here\nexport ELENA_API_BASE_URL=https://elena.sumike.ai\n\n# Optional: Email addresses (if different from defaults)\nexport HANNAH_EMAIL=hannah@sumike.ai\nexport ELENA_EMAIL=elena@sumike.ai\n\n\nOr add to your .env file:\n\nHANNAH_API_KEY=sk-sumike-xxxxx\nELENA_API_KEY=sk-sumike-yyyyy\n\nStep 3: Use the Tools\n\nOnce configured, your agent has access to these tools:\n\nlist_coworkers - List available agents (Hannah, Elena)\nhannah_create_task - Create research task for Hannah\nelena_create_task - Create planning task for Elena\ncheck_task_status - Check status of a task\nget_task_result - Get result of completed task\nhannah_email - Send email request to Hannah\nelena_email - Send email request to Elena\nUsage Examples\nExample 1: Market Research with Hannah (API)\nAgent: \"I need to research the German EV market. Use Hannah.\"\n\n<hannah_create_task>\n  <description>\n    Research the current state of the German electric vehicle market:\n    - Market size and growth trends\n    - Key players and market share\n    - Consumer sentiment and barriers to adoption\n    - Competitive landscape\n  </description>\n  <depth>deep</depth>\n</hannah_create_task>\n\nResult: Task created (task_xyz789).\n\n⏱️ WAIT 2-3 MINUTES before checking status.\n\n[Agent waits 3 minutes]\n\n<check_task_status taskId=\"task_xyz789\">\n\nResult: Task completed! Hannah orchestrated Statista, GWI, and DataForSEO sub-agents.\n\n<get_task_result taskId=\"task_xyz789\">\n\nResult: Comprehensive research report with sources and confidence levels.\n\nExample 2: Project Planning with Elena (API)\nAgent: \"Need to break down a Q2 campaign launch. Use Elena.\"\n\n<elena_create_task>\n  <description>\n    Break down a Q2 product campaign launch into workstreams:\n    - Product: New premium EV model\n    - Target: German market, affluent early adopters\n    - Timeline: Launch April 1st\n    - Budget: €500k\n\n    Provide detailed work breakdown with dependencies and risks.\n  </description>\n</elena_create_task>\n\nResult: Task created (task_abc456).\n\n⏱️ WAIT 2-3 MINUTES - Elena may delegate to Hannah for market research.\n\n[Agent waits 3 minutes]\n\n<check_task_status taskId=\"task_abc456\">\n\nResult: Completed! Elena delegated to Hannah for market data, then integrated findings.\n\n<get_task_result taskId=\"task_abc456\">\n\nResult: Work breakdown with dependencies, risks, and resource needs informed by research.\n\nExample 3: Email-Based Research Request\nAgent: \"Send research request to Hannah via email\"\n\n<hannah_email>\n  <to>hannah@sumike.ai</to>\n  <subject>EV Market Research Request</subject>\n  <body>\n    Hi Hannah,\n\n    I need comprehensive research on the German EV market for Q2 2026:\n    1. Market size and growth projections\n    2. Competitive landscape analysis\n    3. Consumer sentiment (barriers and motivators)\n\n    Please provide sources and confidence levels for all findings.\n\n    Thanks!\n  </body>\n</hannah_email>\n\nResult: Email sent. Hannah replies within 5-10 minutes with research report\n        as email body + attached deliverables (PDF, XLSX).\n\nExample 4: Elena Delegates to Hannah\nAgent: \"Elena, plan the EV campaign launch\"\n\n<elena_create_task>\n  <description>\n    Create project plan for German EV campaign launch:\n    - Product: Premium EV sedan\n    - Launch date: April 1, 2026\n    - Target audience: Affluent professionals, 35-55\n    - Budget: €500k\n  </description>\n</elena_create_task>\n\nWorkflow:\n1. Elena receives task\n2. Elena identifies missing market context\n3. Elena delegates research to Hannah (internal A2A)\n4. Hannah executes research using premium data sources\n5. Elena integrates findings into operational plan\n6. Elena delivers: Work breakdown + dependencies + risk matrix + deliverables\n\nResult: Comprehensive project plan informed by real market data\n\nAgent Personalities & Best Practices\nWorking with Hannah\n\nCharacter:\n\nAnalytical, reflective, opinionated\nValues research integrity over speed\nCalls out weak data honestly\n\"What is actually true?\"\n\nBest Practices:\n\nBe specific about research questions\nSpecify required depth (quick lookup vs deep research)\nAsk for source attribution\nExpect critical assessment, not just data dumps\n\nExample Good Request:\n\n\"Hannah, I need to validate whether 'sustainability' is a real\npurchase driver for EVs in Germany, or if it's post-rationalized.\nGive me data from GWI or Statista if available, and flag if the\ndata quality is weak.\"\n\nWorking with Elena\n\nCharacter:\n\nDirect, pragmatic, grounded\nProtects delivery through realism\nChallenges vague goals\n\"What is actually doable?\"\n\nBest Practices:\n\nProvide clear goals and constraints upfront\nSpecify dependencies you know about\nAsk for risk assessment\nExpect honest pushback on unrealistic plans\n\nExample Good Request:\n\n\"Elena, break down a 6-week campaign launch:\n- Product: New EV model\n- Launch: April 1st\n- Constraint: No dedicated PM resource\n- Concern: Unclear target audience\n\nFlag dependencies and risks.\"\n\nMulti-Agent Coordination\nPattern 1: Sequential (Research → Planning)\nStep 1: Agent requests research from Hannah\nStep 2: Agent receives research findings\nStep 3: Agent requests operational plan from Elena\nStep 4: Agent integrates both outputs\n\nPattern 2: Elena Auto-Delegates\nStep 1: Agent requests project plan from Elena\nStep 2: Elena detects missing market context\nStep 3: Elena automatically delegates to Hannah\nStep 4: Hannah returns research\nStep 5: Elena integrates and delivers complete plan\n\nPattern 3: Parallel Consultation\nStep 1: Agent sends same context to both Hannah and Elena\nStep 2: Hannah returns \"What is true?\" perspective\nStep 3: Elena returns \"What is doable?\" perspective\nStep 4: Agent synthesizes both viewpoints\n\nCost & Rate Limits\nAPI Rate Limits\nHannah: 60 requests/minute\nElena: 60 requests/minute\nMax concurrent requests: 5 per agent\nSokosumi Sub-Agent Costs\n\nWhen Hannah or Elena orchestrate Sokosumi sub-agents:\n\nStatista: ~120 credits\nGWI: ~80 credits\nDataForSEO: ~60 credits\nApify: ~40 credits\n\nHannah and Elena will inform you of estimated costs before executing.\n\n⏱️ Timing Guidance\n\nCRITICAL for AI Agents:\n\nAPI Task Processing\nAfter creating task: Wait at least 2-3 minutes before checking status\nIf still processing: Wait another 2-3 minutes before checking again\nTotal task time: Typically 2-10 minutes depending on complexity\nDO NOT: Poll continuously every few seconds - tasks need time to complete\n\nWhy? Tasks involve:\n\nInitial processing and context loading (30 seconds)\nResearch execution with sub-agents (2-10 minutes - MAIN WAIT TIME)\nResult synthesis and formatting (30 seconds - 1 minute)\nEmail Requests\nTypical response time: 5-10 minutes\nComplex research: Up to 15 minutes\nExpect email reply with attachments\nTask Status States\nStatus\tDescription\tNext Action\nDRAFT\tTask created but not started\tSystem will auto-process\nREADY\tTask queued for processing\tWait 2-3 minutes\nIN_PROGRESS\tAgent is working\tWait 2-3 minutes, then check again\nCOMPLETED\tTask finished\tRetrieve result\nFAILED\tTask failed\tCheck error message\n\nAI Agent: ALWAYS wait 2-3 minutes after creating a task before checking status. Set timeouts of at least 5 minutes for API calls.\n\nAvailable Tools\nlist_coworkers\n\nList available agents (Hannah, Elena).\n\nParameters: None\n\nReturns:\n\ncoworkers: Array of available agents with capabilities and status\ncount: Number of agents found\n\nExample Response:\n\n{\n  \"data\": [\n    {\n      \"id\": \"cow_hannah\",\n      \"name\": \"Hannah Sumi\",\n      \"role\": \"Marketing Research Specialist\",\n      \"email\": \"hannah@sumike.ai\"\n    },\n    {\n      \"id\": \"cow_elena\",\n      \"name\": \"Elena\",\n      \"role\": \"Operations & Project Orchestrator\",\n      \"email\": \"elena@sumike.ai\"\n    }\n  ]\n}\n\nhannah_create_task\n\nCreate a research task for Hannah.\n\nParameters:\n\nname (required): Task title (max 120 chars)\ndescription (optional): Detailed task description with research questions\nstatus (optional): \"DRAFT\" | \"READY\" (default: \"READY\")\n\nReturns:\n\ntaskId: Task identifier (e.g., \"task_xyz789\")\nstatus: Initial task status\nestimatedTime: \"2-10 minutes\"\nmessage: Includes timing guidance\n\n⏱️ IMPORTANT: Wait 2-3 minutes before checking status!\n\nelena_create_task\n\nCreate a planning task for Elena.\n\nParameters:\n\nname (required): Task title (max 120 chars)\ndescription (optional): Detailed planning requirements\nstatus (optional): \"DRAFT\" | \"READY\" (default: \"READY\")\n\nReturns:\n\ntaskId: Task identifier\nstatus: Initial task status\nestimatedTime: \"2-10 minutes\"\nmessage: Includes timing guidance\n\n⏱️ IMPORTANT: Wait 2-3 minutes before checking status! Elena may delegate to Hannah for research.\n\ncheck_task_status\n\nCheck the status of a task.\n\nParameters:\n\ntaskId (required): Task ID from create_task\n\nReturns:\n\nstatus: \"DRAFT\" | \"READY\" | \"IN_PROGRESS\" | \"COMPLETED\" | \"FAILED\"\nhasResult: Whether result is available\nmessage: Status message with timing guidance\n\n⏱️ TIMING: Wait 2-3 minutes after creating task before first check. If still IN_PROGRESS, wait another 2-3 minutes.\n\nget_task_result\n\nGet the result of a completed task.\n\nParameters:\n\ntaskId (required): Task ID from create_task\n\nReturns:\n\nresult: Task result data (research findings or operational plan)\nstatus: Task status (must be \"COMPLETED\")\ncompletedAt: Completion timestamp\ndeliverables: Links to any generated files (PDF, XLSX, PPTX)\n\nNote: Only works for completed tasks. Use check_task_status first to verify completion.\n\nhannah_email\n\nSend email request to Hannah.\n\nParameters:\n\nto (required): Email address (default: hannah@sumike.ai)\nsubject (required): Email subject line\nbody (required): Email body with request details\ncc (optional): CC addresses\nattachments (optional): Attachment file paths\n\nReturns:\n\nstatus: \"sent\"\nmessageId: Email message ID\nestimatedResponse: Estimated response time\nelena_email\n\nSend email request to Elena.\n\nParameters:\n\nto (required): Email address (default: elena@sumike.ai)\nsubject (required): Email subject line\nbody (required): Email body with request details\ncc (optional): CC addresses\nattachments (optional): Attachment file paths\n\nReturns:\n\nstatus: \"sent\"\nmessageId: Email message ID\nestimatedResponse: Estimated response time\ncheck_hannah_status\n\nCheck if Hannah is available.\n\nReturns:\n\navailable: true | false\nresponseTime: Estimated response time\nmessage: Status message\ncheck_elena_status\n\nCheck if Elena is available.\n\nReturns:\n\navailable: true | false\nresponseTime: Estimated response time\nmessage: Status message\nError Handling\nAPI Errors\nError\tCause\tSolution\n401 Unauthorized\tInvalid or missing API key\tSet correct API key in environment\n429 Rate Limited\tExceeded 60 req/min\tWait 60 seconds before retrying\n503 Service Unavailable\tAgent temporarily down\tRetry after 2-3 minutes or use email channel\ntimeout\tRequest took too long\tIncrease timeout for research tasks\nEmail Errors\nError\tCause\tSolution\nEmail not sent\tInvalid email address\tCheck HANNAH_EMAIL / ELENA_EMAIL config\nNo response after 15 min\tComplex research task\tWait longer or check spam folder\nBounce\tEmail delivery failed\tVerify email address and network\nTroubleshooting\n\"API key is missing\"\n\nSolution: Set HANNAH_API_KEY and/or ELENA_API_KEY in environment variables.\n\n\"Hannah/Elena not responding\"\n\nSolution:\n\nCheck agent status using check_hannah_status or check_elena_status\nVerify API endpoint is correct\nTry alternative channel (API → Email or vice versa)\n\"Sokosumi credits exhausted\"\n\nHannah/Elena will inform you:\n\n\"I need to use Statista for this research, which requires 120 credits.\nYour current Sokosumi balance is 50 credits. Please add credits at\nsokosumi.com to continue.\"\n\n\nSolution: Add credits to your Sokosumi account (if using premium data sources)\n\n\"Rate limit exceeded\"\n\nSolution:\n\nReduce request frequency (max 60/minute per agent)\nUse batch requests where possible\nConsider email channel for non-urgent requests\nIntegration Patterns\nPattern A: Research-First Workflow\n1. Your agent identifies need for market data\n2. Call hannah_research with specific questions\n3. Wait for response (2-10 minutes)\n4. Integrate findings into your agent's output\n5. Optionally: Send to Elena for operational planning\n\nPattern B: Planning-First Workflow\n1. Your agent receives project request\n2. Call elena_plan with requirements\n3. Elena auto-delegates research to Hannah if needed\n4. Receive comprehensive plan with market context\n5. Execute or refine based on deliverables\n\nPattern C: Parallel Advisory\n1. Your agent faces strategic decision\n2. Call hannah_research for market reality\n3. Call elena_plan for operational feasibility\n4. Compare responses: \"What's true\" vs \"What's doable\"\n5. Make informed decision with both perspectives\n\nSecurity & Privacy\nAPI Keys\nAPI keys are scoped to specific agents (Hannah or Elena)\nKeys format: sk-sumike-xxxxx\nMaximum 5 active keys per contact\nRotate keys every 90 days recommended\nData Handling\nConversation history stored in PostgreSQL (isolated per agent)\nResearch results cached for 30 days\nEmail attachments stored in Azure File Share\nNo data shared between Hannah and Elena without explicit delegation\nCompliance\nGDPR compliant (data processing agreement available)\nSOC 2 Type II certified infrastructure\nData residency: EU (Azure West Europe)\nAdvanced Configuration\nCustom Endpoints\n# Override default endpoints\nexport HANNAH_API_ENDPOINT=https://custom-hannah.your-domain.com/v1\nexport ELENA_API_ENDPOINT=https://custom-elena.your-domain.com/v1\n\nProxy Configuration\n# Route through proxy\nexport HTTPS_PROXY=https://your-proxy.com:8080\nexport NO_PROXY=localhost,127.0.0.1\n\nTimeout Settings\n# Adjust timeouts (in seconds)\nexport HANNAH_TIMEOUT=300  # 5 minutes for research\nexport ELENA_TIMEOUT=180   # 3 minutes for planning\n\nAPI Reference (Quick Reference)\nHannah API\n\nEndpoint: https://hannah.sumike.ai/v1/chat/completions\n\nAuthentication: Authorization: Bearer sk-sumike-xxxxx\n\nRequest Body:\n\n{\n  \"messages\": [\n    {\"role\": \"user\", \"content\": \"Your research request here\"}\n  ],\n  \"stream\": false\n}\n\n\nResponse:\n\n{\n  \"id\": \"msg_xxx\",\n  \"choices\": [\n    {\n      \"message\": {\n        \"role\": \"assistant\",\n        \"content\": \"Research findings...\"\n      }\n    }\n  ]\n}\n\nElena API\n\nEndpoint: https://elena.sumike.ai/v1/chat/completions\n\nSame format as Hannah API.\n\nResources\nHannah & Elena Homepage: https://sumike.ai\nServiceplan: https://www.serviceplan.com\nSokosumi Marketplace: https://sokosumi.com\nSupport: support@sumike.ai\nSummary for AI Agents\n\nQuick Setup:\n\nObtain API keys from Serviceplan (human task)\nSet HANNAH_API_KEY and ELENA_API_KEY in environment\nUse hannah_research for market research\nUse elena_plan for project planning\nBe patient — research tasks take 2-10 minutes\nIntegrate responses into your agent's workflow\n\nKey Principles:\n\nHannah validates ideas through research integrity (\"What is true?\")\nElena protects delivery through realism (\"What is doable?\")\nBoth agents can orchestrate Sokosumi sub-agents for premium data\nElena auto-delegates to Hannah when market context is needed\nSupport both API and email channels for flexibility\n\nRemember:\n\nResearch depth impacts response time\nSet realistic timeouts (3-5 minutes minimum)\nHannah and Elena are professional coworkers, not assistants\nThey will push back on vague requests or weak assumptions\nCost transparency — they inform you before consuming credits\n\nBuilt by Serviceplan | Powered by Sokosumi\n\nProfessional AI coworkers for marketing research and operations planning"
  },
  "trust": {
    "sourceLabel": "tencent",
    "provenanceUrl": "https://clawhub.ai/Sarthib7/coworker",
    "publisherUrl": "https://clawhub.ai/Sarthib7/coworker",
    "owner": "Sarthib7",
    "version": "1.0.0",
    "license": null,
    "verificationStatus": "Indexed source record"
  },
  "links": {
    "detailUrl": "https://openagent3.xyz/skills/coworker",
    "downloadUrl": "https://openagent3.xyz/downloads/coworker",
    "agentUrl": "https://openagent3.xyz/skills/coworker/agent",
    "manifestUrl": "https://openagent3.xyz/skills/coworker/agent.json",
    "briefUrl": "https://openagent3.xyz/skills/coworker/agent.md"
  }
}