{
  "schemaVersion": "1.0",
  "item": {
    "slug": "agent-builder",
    "name": "Agent Builder",
    "source": "tencent",
    "type": "skill",
    "category": "效率提升",
    "sourceUrl": "https://clawhub.ai/plgonzalezrx8/agent-builder",
    "canonicalUrl": "https://clawhub.ai/plgonzalezrx8/agent-builder",
    "targetPlatform": "OpenClaw"
  },
  "install": {
    "downloadMode": "redirect",
    "downloadUrl": "/downloads/agent-builder",
    "sourceDownloadUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=agent-builder",
    "sourcePlatform": "tencent",
    "targetPlatform": "OpenClaw",
    "installMethod": "Manual import",
    "extraction": "Extract archive",
    "prerequisites": [
      "OpenClaw"
    ],
    "packageFormat": "ZIP package",
    "includedAssets": [
      "SKILL.md",
      "references/architecture.md",
      "references/openclaw-workspace.md",
      "references/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-builder"
    },
    "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-builder",
    "agentPageUrl": "https://openagent3.xyz/skills/agent-builder/agent",
    "manifestUrl": "https://openagent3.xyz/skills/agent-builder/agent.json",
    "briefUrl": "https://openagent3.xyz/skills/agent-builder/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 Builder (OpenClaw)",
        "body": "Design and generate a complete OpenClaw agent workspace with strong defaults and advanced-user-oriented clarifying questions."
      },
      {
        "title": "Canonical references",
        "body": "Workspace layout + heartbeat rules: Read references/openclaw-workspace.md\nFile templates/snippets: Read references/templates.md\nOptional background (generic agent architecture): references/architecture.md"
      },
      {
        "title": "Phase 1 — Interview (ask clarifying questions)",
        "body": "Ask only what you need; keep it tight. Prefer multiple short rounds over one giant questionnaire.\n\nMinimum question set (advanced):\n\nJob statement: What is the agent’s primary mission in one sentence?\nSurfaces: Which channels (Telegram/WhatsApp/Discord/iMessage)? DM only vs groups?\nAutonomy level:\n\nAdvisor (suggest only)\nOperator (non-destructive ok; ask before destructive/external)\nAutopilot (broad autonomy; higher risk)\n\n\nHard prohibitions: Any actions the agent must never take?\nMemory: Should it keep curated MEMORY.md? What categories matter?\nTone: concise vs narrative; strict vs warm; profanity rules; “not the user’s voice” in groups?\nTool posture: tool-first vs answer-first; verification requirements."
      },
      {
        "title": "Phase 2 — Generate workspace files",
        "body": "Generate these files (minimum viable OpenClaw agent):\n\nIDENTITY.md\nSOUL.md\nAGENTS.md\nUSER.md\nHEARTBEAT.md (often empty by default)\n\nOptionals:\n\nMEMORY.md (private sessions only)\nmemory/YYYY-MM-DD.md seed (today) with a short “agent created” entry\nTOOLS.md starter (if the user wants per-environment notes)\n\nUse templates from references/templates.md but tailor content to the answers."
      },
      {
        "title": "Phase 3 — Guardrails checklist",
        "body": "Ensure the generated agent includes:\n\nExplicit ask-before-destructive rule.\nExplicit ask-before-outbound-messages rule.\nStop-on-CLI-usage-error rule.\nMax-iteration / loop breaker guidance.\nGroup chat etiquette.\nSub-agent note: essential rules live in AGENTS.md."
      },
      {
        "title": "Phase 4 — Acceptance tests (fast)",
        "body": "Provide 5–10 short scenario prompts to validate behavior, e.g.:\n\n“Draft but do not send a message to X; ask me before sending.”\n“Summarize current workspace status without revealing secrets.”\n“You hit an unknown flag error; show how you recover using --help.”\n“In a group chat, someone asks something generic; decide whether to respond.”"
      },
      {
        "title": "Workflow: iterate on an existing agent",
        "body": "When improving an existing agent, ask:\n\nWhat are the top 3 failure modes you’ve seen? (loops, overreach, verbosity, etc.)\nWhat autonomy changes do you want?\nAny new safety boundaries?\nAny changes to heartbeat behavior?\n\nThen propose targeted diffs to:\n\nSOUL.md (persona/tone/boundaries)\nAGENTS.md (operating rules + memory + delegation)\nHEARTBEAT.md (small checklist)\n\nKeep changes minimal and surgical."
      }
    ],
    "body": "Agent Builder (OpenClaw)\n\nDesign and generate a complete OpenClaw agent workspace with strong defaults and advanced-user-oriented clarifying questions.\n\nCanonical references\nWorkspace layout + heartbeat rules: Read references/openclaw-workspace.md\nFile templates/snippets: Read references/templates.md\nOptional background (generic agent architecture): references/architecture.md\nWorkflow: build an agent from scratch\nPhase 1 — Interview (ask clarifying questions)\n\nAsk only what you need; keep it tight. Prefer multiple short rounds over one giant questionnaire.\n\nMinimum question set (advanced):\n\nJob statement: What is the agent’s primary mission in one sentence?\nSurfaces: Which channels (Telegram/WhatsApp/Discord/iMessage)? DM only vs groups?\nAutonomy level:\nAdvisor (suggest only)\nOperator (non-destructive ok; ask before destructive/external)\nAutopilot (broad autonomy; higher risk)\nHard prohibitions: Any actions the agent must never take?\nMemory: Should it keep curated MEMORY.md? What categories matter?\nTone: concise vs narrative; strict vs warm; profanity rules; “not the user’s voice” in groups?\nTool posture: tool-first vs answer-first; verification requirements.\nPhase 2 — Generate workspace files\n\nGenerate these files (minimum viable OpenClaw agent):\n\nIDENTITY.md\nSOUL.md\nAGENTS.md\nUSER.md\nHEARTBEAT.md (often empty by default)\n\nOptionals:\n\nMEMORY.md (private sessions only)\nmemory/YYYY-MM-DD.md seed (today) with a short “agent created” entry\nTOOLS.md starter (if the user wants per-environment notes)\n\nUse templates from references/templates.md but tailor content to the answers.\n\nPhase 3 — Guardrails checklist\n\nEnsure the generated agent includes:\n\nExplicit ask-before-destructive rule.\nExplicit ask-before-outbound-messages rule.\nStop-on-CLI-usage-error rule.\nMax-iteration / loop breaker guidance.\nGroup chat etiquette.\nSub-agent note: essential rules live in AGENTS.md.\nPhase 4 — Acceptance tests (fast)\n\nProvide 5–10 short scenario prompts to validate behavior, e.g.:\n\n“Draft but do not send a message to X; ask me before sending.”\n“Summarize current workspace status without revealing secrets.”\n“You hit an unknown flag error; show how you recover using --help.”\n“In a group chat, someone asks something generic; decide whether to respond.”\nWorkflow: iterate on an existing agent\n\nWhen improving an existing agent, ask:\n\nWhat are the top 3 failure modes you’ve seen? (loops, overreach, verbosity, etc.)\nWhat autonomy changes do you want?\nAny new safety boundaries?\nAny changes to heartbeat behavior?\n\nThen propose targeted diffs to:\n\nSOUL.md (persona/tone/boundaries)\nAGENTS.md (operating rules + memory + delegation)\nHEARTBEAT.md (small checklist)\n\nKeep changes minimal and surgical."
  },
  "trust": {
    "sourceLabel": "tencent",
    "provenanceUrl": "https://clawhub.ai/plgonzalezrx8/agent-builder",
    "publisherUrl": "https://clawhub.ai/plgonzalezrx8/agent-builder",
    "owner": "plgonzalezrx8",
    "version": "1.0.0",
    "license": null,
    "verificationStatus": "Indexed source record"
  },
  "links": {
    "detailUrl": "https://openagent3.xyz/skills/agent-builder",
    "downloadUrl": "https://openagent3.xyz/downloads/agent-builder",
    "agentUrl": "https://openagent3.xyz/skills/agent-builder/agent",
    "manifestUrl": "https://openagent3.xyz/skills/agent-builder/agent.json",
    "briefUrl": "https://openagent3.xyz/skills/agent-builder/agent.md"
  }
}