{ "schemaVersion": "1.0", "item": { "slug": "council-builder", "name": "Council Builder", "source": "tencent", "type": "skill", "category": "AI 智能", "sourceUrl": "https://clawhub.ai/Abdullah4AI/council-builder", "canonicalUrl": "https://clawhub.ai/Abdullah4AI/council-builder", "targetPlatform": "OpenClaw" }, "install": { "downloadMode": "redirect", "downloadUrl": "/downloads/council-builder", "sourceDownloadUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=council-builder", "sourcePlatform": "tencent", "targetPlatform": "OpenClaw", "installMethod": "Manual import", "extraction": "Extract archive", "prerequisites": [ "OpenClaw" ], "packageFormat": "ZIP package", "includedAssets": [ "SKILL.md", "assets/ADAPTIVE-ROUTING-LEARNING-TEMPLATE.md", "assets/AGENT-AGENTS-TEMPLATE.md", "assets/LEARNING-METRICS-TEMPLATE.json", "assets/LEARNINGS-TEMPLATE.md", "assets/ROOT-AGENTS-TEMPLATE.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": "2026-04-30T16:55:25.780Z", "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/council-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/council-builder", "agentPageUrl": "https://openagent3.xyz/skills/council-builder/agent", "manifestUrl": "https://openagent3.xyz/skills/council-builder/agent.json", "briefUrl": "https://openagent3.xyz/skills/council-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": "Council Builder", "body": "Build a team of specialized AI agent personas tailored to the user's actual needs. Each agent gets a distinct personality, self-improvement capability, and clear coordination rules." }, { "title": "Phase 1: Discovery", "body": "Interview the user to understand their world. Ask in batches of 2-3 questions max.\n\nRound 1 - Identity:\n\nWhat do you do? (profession, main activities, industry)\nWhat tools and platforms do you use daily?\n\nRound 2 - Pain Points:\n\nWhat tasks eat most of your time?\nWhere do you feel you need the most help?\n\nRound 3 - Preferences:\n\nWhat language(s) do you work in? (for agent communication style)\nAny specific domains you want covered? (coding, content, finance, research, scheduling, etc.)\n\nOptional - History Analysis:\nIf the user has existing OpenClaw history, scan it for patterns:\n\nCheck memory/ files for recurring tasks\nCheck existing workspace structure for active projects\nCheck installed skills for current capabilities\n\nDo NOT proceed to Phase 2 until confident you understand the user's needs. Ask follow-up questions if anything is unclear." }, { "title": "Phase 2: Planning", "body": "Based on discovery, design the council:\n\nDetermine agent count: 3-7 agents. Fewer is better. Each agent must earn its existence.\nDefine each agent: Name, role, specialties, personality angle\nMap coordination: Which agents feed data to which\nPresent the plan to the user in a clear table:\n\n| Agent | Role | Specialties | Personality |\n|-------|------|-------------|-------------|\n| [Name] | [One-line role] | [Key areas] | [Personality angle] |\n\nGet explicit approval before building. Allow adjustments.\n\nNaming agents:\n\nGive them memorable, short names (not generic like \"Agent 1\")\nNames should hint at their role but feel like characters\nCan be inspired by any theme the user likes, or choose strong standalone names\nSee references/example-councils.md for naming patterns and complete council examples across different industries" }, { "title": "Phase 3: Building", "body": "Run the initialization script first to create the directory skeleton:\n\n./scripts/init-council.sh ...\n\nThen, for each approved agent, populate the files. Read references/soul-philosophy.md before writing any SOUL.md.\n\nDirectory structure per agent:\n\nagents/[agent-name]/\n├── SOUL.md # Personality, role, rules (see soul-philosophy.md)\n├── AGENTS.md # Agent-specific coordination rules\n├── memory/ # Agent's memory directory\n├── .learnings/ # Self-improvement logs\n│ ├── LEARNINGS.md\n│ ├── ERRORS.md\n│ └── FEATURE_REQUESTS.md\n└── [workspace dirs] # Role-specific output directories\n\nFor each agent's SOUL.md:\n\nRead references/soul-philosophy.md for the writing guide\nRead assets/SOUL-TEMPLATE.md for the structure\nCustomize deeply for this agent's role and personality\nEvery SOUL must be unique. No copy-paste between agents.\n\nFor each agent's AGENTS.md:\n\nUse assets/AGENT-AGENTS-TEMPLATE.md as base\nDefine what this agent reads from and writes to\nDefine handoff rules with other agents\n\nFor .learnings/ files:\n\nCopy structure from assets/LEARNINGS-TEMPLATE.md\nInitialize empty log files\n\nFor the root AGENTS.md:\n\nUse assets/ROOT-AGENTS-TEMPLATE.md as base\nCreate the routing table for all agents\nDefine file coordination map\nSet up enforcement rules\nAdd adaptive model routing thresholds (Fast, Think, Deep, Strategic)" }, { "title": "Phase 4: Adaptive Routing Setup", "body": "Read references/adaptive-routing.md.\n\nSet up an adaptive routing section in root AGENTS.md:\n\nDefault to Fast\nEscalation thresholds for Think, Deep, Strategic\nDe-escalation rule back to Fast after heavy reasoning\nHigh-tier model rate-limit fallback behavior\n\nAlso create visual architecture doc:\n\ndocs/architecture/ADAPTIVE-ROUTING-LEARNING.md using assets/ADAPTIVE-ROUTING-LEARNING-TEMPLATE.md" }, { "title": "Phase 5: Self-Improvement Setup", "body": "Read references/self-improvement.md for the complete system.\n\nEach agent gets built-in self-improvement:\n\n.learnings/ directory with proper templates\nDetection triggers in SOUL.md (corrections, errors, gaps)\nPromotion rules (learning → SOUL.md / AGENTS.md / TOOLS.md)\nCross-agent learning sharing via shared/learnings/CROSS-AGENT.md\nPeriodic review instructions\nWeekly learning metrics file at memory/learning-metrics.json (use assets/LEARNING-METRICS-TEMPLATE.json)" }, { "title": "Phase 6: Verification", "body": "After building everything:\n\nList all created files for the user\nShow the routing table\nShow the coordination map\nConfirm everything is in place" }, { "title": "Phase 7: Expansion (On-Demand)", "body": "When the user asks to add, modify, or remove agents:\n\nAdding an agent:\n\nMini-discovery: What does this agent need to do?\nCreate full agent structure (same as Phase 3)\nUpdate root AGENTS.md routing table\nUpdate coordination map\n\nModifying an agent:\n\nRead the current SOUL.md\nApply changes while preserving personality consistency\nUpdate related coordination rules if needed\n\nRemoving an agent:\n\nAsk for confirmation\nReassign the agent's responsibilities to other agents\nUpdate routing table and coordination map\nMove agent files to trash (never delete)" }, { "title": "Key Principles", "body": "Each agent is a character, not a template. Different personality, different voice, different strengths. If two agents sound the same, one shouldn't exist.\n\n\nNo corporate language in any SOUL. See references/soul-philosophy.md. This is non-negotiable.\n\n\nSelf-improvement is mandatory. Every agent logs mistakes and learns. See references/self-improvement.md.\n\n\nCoordination through files. Agents communicate via shared directories, not direct messaging. Each agent has clear read/write boundaries.\n\n\nBrevity in everything. SOULs, AGENTS files, templates. Respect the context window.\n\n\nThe user's main assistant is the coordinator. It routes tasks, not the agents themselves.\n\n\nLanguage-adaptive. Write SOULs in whatever language the user works in. Arabic, English, bilingual, whatever fits their world.\n\n\nAdaptive routing by default. Every generated council should include Fast/Think/Deep/Strategic model routing thresholds.\n\n\nMetrics over vibes. Weekly learning review must be measured in memory/learning-metrics.json.\n\n\nArchitecture must be visual. Generate a concise architecture doc at docs/architecture/ADAPTIVE-ROUTING-LEARNING.md for training and onboarding." } ], "body": "Council Builder\n\nBuild a team of specialized AI agent personas tailored to the user's actual needs. Each agent gets a distinct personality, self-improvement capability, and clear coordination rules.\n\nWorkflow\nPhase 1: Discovery\n\nInterview the user to understand their world. Ask in batches of 2-3 questions max.\n\nRound 1 - Identity:\n\nWhat do you do? (profession, main activities, industry)\nWhat tools and platforms do you use daily?\n\nRound 2 - Pain Points:\n\nWhat tasks eat most of your time?\nWhere do you feel you need the most help?\n\nRound 3 - Preferences:\n\nWhat language(s) do you work in? (for agent communication style)\nAny specific domains you want covered? (coding, content, finance, research, scheduling, etc.)\n\nOptional - History Analysis: If the user has existing OpenClaw history, scan it for patterns:\n\nCheck memory/ files for recurring tasks\nCheck existing workspace structure for active projects\nCheck installed skills for current capabilities\n\nDo NOT proceed to Phase 2 until confident you understand the user's needs. Ask follow-up questions if anything is unclear.\n\nPhase 2: Planning\n\nBased on discovery, design the council:\n\nDetermine agent count: 3-7 agents. Fewer is better. Each agent must earn its existence.\nDefine each agent: Name, role, specialties, personality angle\nMap coordination: Which agents feed data to which\nPresent the plan to the user in a clear table:\n| Agent | Role | Specialties | Personality |\n|-------|------|-------------|-------------|\n| [Name] | [One-line role] | [Key areas] | [Personality angle] |\n\nGet explicit approval before building. Allow adjustments.\n\nNaming agents:\n\nGive them memorable, short names (not generic like \"Agent 1\")\nNames should hint at their role but feel like characters\nCan be inspired by any theme the user likes, or choose strong standalone names\nSee references/example-councils.md for naming patterns and complete council examples across different industries\nPhase 3: Building\n\nRun the initialization script first to create the directory skeleton:\n\n./scripts/init-council.sh ...\n\n\nThen, for each approved agent, populate the files. Read references/soul-philosophy.md before writing any SOUL.md.\n\nDirectory structure per agent:\n\nagents/[agent-name]/\n├── SOUL.md # Personality, role, rules (see soul-philosophy.md)\n├── AGENTS.md # Agent-specific coordination rules\n├── memory/ # Agent's memory directory\n├── .learnings/ # Self-improvement logs\n│ ├── LEARNINGS.md\n│ ├── ERRORS.md\n│ └── FEATURE_REQUESTS.md\n└── [workspace dirs] # Role-specific output directories\n\n\nFor each agent's SOUL.md:\n\nRead references/soul-philosophy.md for the writing guide\nRead assets/SOUL-TEMPLATE.md for the structure\nCustomize deeply for this agent's role and personality\nEvery SOUL must be unique. No copy-paste between agents.\n\nFor each agent's AGENTS.md:\n\nUse assets/AGENT-AGENTS-TEMPLATE.md as base\nDefine what this agent reads from and writes to\nDefine handoff rules with other agents\n\nFor .learnings/ files:\n\nCopy structure from assets/LEARNINGS-TEMPLATE.md\nInitialize empty log files\n\nFor the root AGENTS.md:\n\nUse assets/ROOT-AGENTS-TEMPLATE.md as base\nCreate the routing table for all agents\nDefine file coordination map\nSet up enforcement rules\nAdd adaptive model routing thresholds (Fast, Think, Deep, Strategic)\nPhase 4: Adaptive Routing Setup\n\nRead references/adaptive-routing.md.\n\nSet up an adaptive routing section in root AGENTS.md:\n\nDefault to Fast\nEscalation thresholds for Think, Deep, Strategic\nDe-escalation rule back to Fast after heavy reasoning\nHigh-tier model rate-limit fallback behavior\n\nAlso create visual architecture doc:\n\ndocs/architecture/ADAPTIVE-ROUTING-LEARNING.md using assets/ADAPTIVE-ROUTING-LEARNING-TEMPLATE.md\nPhase 5: Self-Improvement Setup\n\nRead references/self-improvement.md for the complete system.\n\nEach agent gets built-in self-improvement:\n\n.learnings/ directory with proper templates\nDetection triggers in SOUL.md (corrections, errors, gaps)\nPromotion rules (learning → SOUL.md / AGENTS.md / TOOLS.md)\nCross-agent learning sharing via shared/learnings/CROSS-AGENT.md\nPeriodic review instructions\nWeekly learning metrics file at memory/learning-metrics.json (use assets/LEARNING-METRICS-TEMPLATE.json)\nPhase 6: Verification\n\nAfter building everything:\n\nList all created files for the user\nShow the routing table\nShow the coordination map\nConfirm everything is in place\nPhase 7: Expansion (On-Demand)\n\nWhen the user asks to add, modify, or remove agents:\n\nAdding an agent:\n\nMini-discovery: What does this agent need to do?\nCreate full agent structure (same as Phase 3)\nUpdate root AGENTS.md routing table\nUpdate coordination map\n\nModifying an agent:\n\nRead the current SOUL.md\nApply changes while preserving personality consistency\nUpdate related coordination rules if needed\n\nRemoving an agent:\n\nAsk for confirmation\nReassign the agent's responsibilities to other agents\nUpdate routing table and coordination map\nMove agent files to trash (never delete)\nKey Principles\n\nEach agent is a character, not a template. Different personality, different voice, different strengths. If two agents sound the same, one shouldn't exist.\n\nNo corporate language in any SOUL. See references/soul-philosophy.md. This is non-negotiable.\n\nSelf-improvement is mandatory. Every agent logs mistakes and learns. See references/self-improvement.md.\n\nCoordination through files. Agents communicate via shared directories, not direct messaging. Each agent has clear read/write boundaries.\n\nBrevity in everything. SOULs, AGENTS files, templates. Respect the context window.\n\nThe user's main assistant is the coordinator. It routes tasks, not the agents themselves.\n\nLanguage-adaptive. Write SOULs in whatever language the user works in. Arabic, English, bilingual, whatever fits their world.\n\nAdaptive routing by default. Every generated council should include Fast/Think/Deep/Strategic model routing thresholds.\n\nMetrics over vibes. Weekly learning review must be measured in memory/learning-metrics.json.\n\nArchitecture must be visual. Generate a concise architecture doc at docs/architecture/ADAPTIVE-ROUTING-LEARNING.md for training and onboarding." }, "trust": { "sourceLabel": "tencent", "provenanceUrl": "https://clawhub.ai/Abdullah4AI/council-builder", "publisherUrl": "https://clawhub.ai/Abdullah4AI/council-builder", "owner": "Abdullah4AI", "version": "1.1.1", "license": null, "verificationStatus": "Indexed source record" }, "links": { "detailUrl": "https://openagent3.xyz/skills/council-builder", "downloadUrl": "https://openagent3.xyz/downloads/council-builder", "agentUrl": "https://openagent3.xyz/skills/council-builder/agent", "manifestUrl": "https://openagent3.xyz/skills/council-builder/agent.json", "briefUrl": "https://openagent3.xyz/skills/council-builder/agent.md" } }