# Send Team Builder to your agent
Hand the extracted package to your coding agent with a concrete install brief instead of figuring it out manually.
## Fast path
- 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.
## Suggested prompts
### New install

```text
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.
```
### Upgrade existing

```text
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.
```
## Machine-readable fields
```json
{
  "schemaVersion": "1.0",
  "item": {
    "slug": "team-builder",
    "name": "Team Builder",
    "source": "tencent",
    "type": "skill",
    "category": "效率提升",
    "sourceUrl": "https://clawhub.ai/beyound87/team-builder",
    "canonicalUrl": "https://clawhub.ai/beyound87/team-builder",
    "targetPlatform": "OpenClaw"
  },
  "install": {
    "downloadUrl": "/downloads/team-builder",
    "sourceDownloadUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=team-builder",
    "sourcePlatform": "tencent",
    "targetPlatform": "OpenClaw",
    "packageFormat": "ZIP package",
    "primaryDoc": "SKILL.md",
    "includedAssets": [
      "references/agent-refs/chief-of-staff/dashboard-template.md",
      "references/agent-refs/chief-of-staff/strategy-methodology.md",
      "references/agent-refs/content-chief/methodology.md",
      "references/agent-refs/data-analyst/methodology.md",
      "references/agent-refs/fullstack-dev/coding-behavior-fallback.md",
      "references/agent-refs/fullstack-dev/deep-dive-protocol.md"
    ],
    "downloadMode": "redirect",
    "sourceHealth": {
      "source": "tencent",
      "slug": "team-builder",
      "status": "healthy",
      "reason": "direct_download_ok",
      "recommendedAction": "download",
      "checkedAt": "2026-05-01T09:55:42.381Z",
      "expiresAt": "2026-05-08T09:55:42.381Z",
      "httpStatus": 200,
      "finalUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=team-builder",
      "contentType": "application/zip",
      "probeMethod": "head",
      "details": {
        "probeUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=team-builder",
        "contentDisposition": "attachment; filename=\"team-builder-3.0.0.zip\"",
        "redirectLocation": null,
        "bodySnippet": null,
        "slug": "team-builder"
      },
      "scope": "item",
      "summary": "Item download looks usable.",
      "detail": "Yavira can redirect you to the upstream package for this item.",
      "primaryActionLabel": "Download for OpenClaw",
      "primaryActionHref": "/downloads/team-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."
      ]
    }
  },
  "links": {
    "detailUrl": "https://openagent3.xyz/skills/team-builder",
    "downloadUrl": "https://openagent3.xyz/downloads/team-builder",
    "agentUrl": "https://openagent3.xyz/skills/team-builder/agent",
    "manifestUrl": "https://openagent3.xyz/skills/team-builder/agent.json",
    "briefUrl": "https://openagent3.xyz/skills/team-builder/agent.md"
  }
}
```
## Documentation

### Team Builder

Deploy a 7-agent SaaS growth team on OpenClaw in one shot.

### System Impact & Prerequisites

Read before running. This skill creates files and modifies system config.

### What it creates

A new workspace directory with ~40 files (agent configs, shared knowledge, inboxes, kanban)
apply-config.js -- script that modifies ~/.openclaw/openclaw.json (adds agents, bindings, agentToAgent config). Auto-backs up before writing.
create-crons.ps1 / create-crons.sh -- scripts that create cron jobs via openclaw cron add
After running these scripts you must restart the gateway (openclaw gateway restart)

### What it does NOT do automatically

Does not modify openclaw.json directly -- you run apply-config.js yourself
Does not create cron jobs directly -- you run the cron script yourself
Does not restart the gateway -- you do that manually

### Optional: Telegram

If you provide bot tokens during setup, apply-config.js will also add Telegram account configs and bindings
Requires: Telegram bot tokens from @BotFather, your Telegram user ID
Requires: network access to Telegram API (proxy configurable)

### Optional: ACP / Claude Code

The fullstack-dev agent is configured to spawn Claude Code via ACP for complex coding tasks
Requires: ACP-compatible coding agent configured in your OpenClaw environment
No extra setup needed if you don't use this feature

### Credentials involved

Telegram bot tokens (optional) -- stored in openclaw.json, used for agent-to-Telegram binding
Model API keys -- must already be configured in your OpenClaw model providers (not handled by this skill)

### Recommended

Review generated apply-config.js before running
Check the backup of openclaw.json after running
Test with 2-3 agents before enabling all cron jobs

### Team Architecture

Default 7-agent SaaS growth team (customizable to 2-10 agents):

CEO
 |-- Chief of Staff (dispatch + strategy + efficiency)
 |-- Data Analyst (data + user research)
 |-- Growth Lead (GEO + SEO + community + social media)
 |-- Content Chief (strategy + writing + copywriting + i18n)
 |-- Intel Analyst (competitor monitoring + market trends)
 |-- Product Lead (product management + tech architecture)
 |-- Fullstack Dev (full-stack dev + ops, spawns Claude Code with role-based prompts)

### Multi-Team Support

One OpenClaw instance can run multiple teams:

node <skill-dir>/scripts/deploy.js                  # default team
node <skill-dir>/scripts/deploy.js --team alpha      # named team "alpha"
node <skill-dir>/scripts/deploy.js --team beta       # named team "beta"

Named teams use prefixed agent IDs (alpha-chief-of-staff, beta-growth-lead) to avoid conflicts. Each team gets its own workspace subdirectory.

### Flexible Team Size

The wizard lets you select 2-10 agents from the available roles. Skip roles you don't need. The 7-agent default covers most SaaS scenarios, but you can run leaner (3-4 agents) or expand with custom roles.

### Model Auto-Detection

The wizard scans your openclaw.json for registered model providers and auto-suggests models by role type:

Role TypeBest ForAuto-detect PatternThinkingStrategic roles (chief, growth, content, product)/glm-5|opus|o1|deepthink/iExecutionOperational roles (data, intel, fullstack)/glm-4|sonnet|gpt-4/iFastLightweight tasks/flash|haiku|mini/i

You can always override with manual model IDs.

### Step 1: Collect Configuration

Ask the user for these inputs (use defaults if not provided):

ParameterDefaultDescriptionTeam nameAlpha TeamUsed in all docs and configsWorkspace dir~/.openclaw/workspace-teamShared workspace rootTimezoneAsia/ShanghaiFor cron schedulesMorning brief hour8Chief's morning reportEvening brief hour18Chief's evening reportThinking modelzai/glm-5For strategic rolesExecution modelzai/glm-4.7For execution rolesCEO titleBossHow agents address the CEO

Optional: Telegram user ID, proxy, and 7 bot tokens.

### Step 2: Run Deploy Script

node <skill-dir>/scripts/deploy.js

Interactive -- asks all questions from Step 1, generates the full workspace.

### Step 3: Apply Config

node <workspace-dir>/apply-config.js

Adds agents to openclaw.json, preserving existing config.

### Step 4: Create Cron Jobs

# Windows
powershell <workspace-dir>/create-crons.ps1

# Linux/Mac
bash <workspace-dir>/create-crons.sh

### Step 5: Restart Gateway

openclaw gateway restart

### Step 6: Fill Business Info

User must edit:

shared/decisions/active.md -- strategy, priorities
shared/products/_index.md -- products overview (≤5 lines per product: URL, code path, positioning, tech, status). Detailed info goes in each product's overview.md.
shared/knowledge/competitor-map.md -- competitor analysis
shared/knowledge/tech-standards.md -- coding standards

### Step 7: Trigger Deep Dive Scans

After filling in products with code directories, tell product-lead to trigger Deep Dive scans:

Product-lead sends scan requests to fullstack-dev via inbox
Fullstack-dev enters each project directory and generates knowledge files
Product-lead reviews the generated files for completeness
All agents now have deep project understanding for informed decisions

### Cron Schedule

OffsetAgentTaskFrequencyH-1Data AnalystData + user feedbackDailyH-1Intel AnalystCompetitor scanMon/Wed/FriHChief of StaffMorning brief (announced)DailyH+1Growth LeadGEO + SEO + communityDailyH+1Content ChiefWeekly content planMondayH+10Chief of StaffEvening brief (announced)Daily

(H = morning brief hour)

### Generated File Structure

<workspace>/
├── AGENTS.md, SOUL.md, USER.md  (auto-injected)
├── apply-config.js, create-crons.ps1/.sh, README.md
├── agents/<7 agent dirs>/       (SOUL.md + MEMORY.md + memory/)
└── shared/
    ├── briefings/, decisions/, inbox/ (v2: with status tracking)
    ├── status/team-dashboard.md     (chief-of-staff maintains, all agents read first)
    ├── data/                        (public data pool, data-analyst writes, all read)
    ├── kanban/, knowledge/
    └── products/
        ├── _index.md                (product matrix overview)
        ├── _template/               (knowledge directory template)
        └── {product}/               (per-product knowledge, up to 20 files)
            ├── overview.md, architecture.md, database.md, api.md, routes.md
            ├── models.md, services.md, frontend.md, auth.md, integrations.md
            ├── jobs-events.md, config-env.md, dependencies.md, devops.md
            ├── test-coverage.md, tech-debt.md, domain-flows.md, data-flow.md
            ├── i18n.md, changelog.md, notes.md

### Knowledge Governance

Each shared knowledge file has a designated owner. Only the owner agent updates it; others read only.

FileOwnerUpdate Triggergeo-playbook.mdgrowth-leadAfter GEO experiments/discoveriesseo-playbook.mdgrowth-leadAfter SEO experimentscompetitor-map.mdintel-analystAfter each competitor scancontent-guidelines.mdcontent-chiefAfter proven writing patternsuser-personas.mddata-analystAfter new user insightstech-standards.mdproduct-leadAfter architecture decisions

### Update Protocol

When updating a knowledge file, the owner must:

Add a dated entry at the top: ## [YYYY-MM-DD] <what changed>
Include the reason and data evidence
Never delete existing entries without CEO approval (append, don't replace)

### Chief of Staff Governance

The chief-of-staff monitors knowledge file health during weekly reviews:

Are files being updated regularly?
Any conflicting information between files?
Any stale entries that should be archived?

### Self-Evolution Pattern

Agents improve their own strategies over time through a feedback loop:

1. Execute task (cron or inbox triggered)
2. Collect results (data, metrics, outcomes)
3. Analyze: what worked vs what didn't
4. Update knowledge files with proven strategies (with evidence)
5. Next execution reads updated knowledge → better performance

This is NOT the agent randomly changing rules. Updates must be:

Data-driven: backed by metrics or concrete outcomes
Incremental: append new findings, don't rewrite everything
Traceable: dated with evidence so others can verify

### What Agents Can Self-Update

Their own knowledge files (per ownership table above)
Their own MEMORY.md (lessons learned, decisions)
shared/data/ outputs (data-analyst only)

### What Requires CEO Approval

shared/decisions/active.md (strategy changes)
Adding/removing agents or changing team architecture
External publishing or spending decisions

### Public Data Layer

The shared/data/ directory serves as a read-only data pool for all agents:

data-analyst writes: daily metrics, user feedback summaries, anomaly alerts
All agents read: to inform their own decisions
Format: structured markdown or JSON, dated filenames (e.g., metrics-2026-03-01.md)
Retention: keep 30 days, archive older files

### Project Deep Dive — Code Scanning

Agents can deeply understand each SaaS product through automated code scanning. This is critical — without deep project knowledge, all team decisions are surface-level.

### How It Works

CEO adds a product to shared/products/_index.md (name, URL, code directory, tech stack)
Product Lead triggers a Deep Dive scan by messaging Fullstack Dev via inbox
Fullstack Dev enters the project directory (read-only) and scans the codebase
Knowledge files are generated in shared/products/{product}/
All agents consume these files via manifest-based lazy loading (never read all at once)

### Manifest-Based Lazy Loading (MANDATORY)

Each product directory includes a manifest.json (~200 tokens) that lists all files with one-line summaries and a taskFileMap mapping task types to relevant files.

Agent workflow:

Read _index.md → identify which product
Read {product}/manifest.json → see all files + summaries (~200 tokens)
Based on taskFileMap or summaries, read only 1-3 relevant files
Never read more than 5 product files per session

Why: With 15+ products × 20 files each, full loading = 40K+ tokens per product. Manifest loading = 200 tokens + only what's needed.

Fullstack Dev MUST regenerate manifest.json after every scan (L0-L4). Template in _template/manifest.json.

### Manifest Quality Standards

摘要不能为了省 token 丢掉关键信息。每条摘要须满足：

核心文件（database/models/services/routes/integrations）：50-130字，列出关键实体名/数量/域名
中等文件（auth/frontend/commands/config）：30-80字，点明方案和范围
轻量文件（changelog/notes/metrics）：可以短（<20字）
taskFileMap：必须覆盖该产品的所有核心业务场景（不少于8个映射）
codeStats：必须包含文件数、行数、模型数、表数等量化指标

### Product Knowledge Directory

Each product gets a knowledge directory with up to 20 files + manifest:

shared/products/{product}/
├── manifest.json        ← **INDEX** (~200 tokens): file list, summaries, taskFileMap
├── overview.md          ← Product positioning (from _index.md)
├── architecture.md      ← System architecture, tech stack, design patterns, layering
├── database.md          ← Full table schema, relationships, indexes, migrations
├── api.md               ← API endpoints, params, auth, versioning
├── routes.md            ← Complete route table (Web + API + Console)
├── models.md            ← ORM relationships, scopes, accessors, observers
├── services.md          ← Business logic, state machines, workflows, validation
├── frontend.md          ← Component tree, page routing, state management
├── auth.md              ← Auth scheme, roles/permissions matrix, OAuth
├── integrations.md      ← Third-party: payment/email/SMS/storage/CDN/analytics
├── jobs-events.md       ← Queue jobs, event listeners, scheduled tasks, notifications
├── config-env.md        ← Environment variables, feature flags, cache strategy
├── dependencies.md      ← Key dependencies, custom packages, vulnerabilities
├── devops.md            ← Deployment, CI/CD, Docker, monitoring, logging
├── test-coverage.md     ← Test strategy, coverage, weak spots
├── tech-debt.md         ← TODO/FIXME/HACK inventory, dead code, complexity hotspots
├── domain-flows.md      ← Core user journeys, domain boundaries, module coupling
├── data-flow.md         ← Data lifecycle: external → import → process → store → output
├── i18n.md              ← Internationalization, language coverage
├── changelog.md         ← Scan diff log (what changed between scans)
└── notes.md             ← Agent discoveries, gotchas, implicit rules

### Scan Levels

LevelScopeWhenOutputL0 SnapshotSurface: directory tree, packages, envFirst onboardarchitecture, dependencies, config-envL1 SkeletonStructure: DB, routes, models, componentsFirst onboarddatabase, routes, api, models, frontendL2 Deep DiveLogic: services, auth, jobs, integrationsOn-demand per moduleservices, auth, jobs-events, integrations, domain-flows, data-flowL3 Health CheckQuality: tech debt, tests, securityPeriodic / pre-releasetech-debt, test-coverage, devopsL4 IncrementalDelta: git diff → update affected filesAfter code changeschangelog + targeted updates

### Content Standards

Knowledge files capture not just WHAT exists but WHY:

Design decisions: Why this approach was chosen
Implicit business rules: Logic buried in code (e.g., "orders auto-cancel after 72h")
Gotchas: What breaks if you touch this module carelessly
Cross-module coupling: Where changing A silently breaks B
Performance hotspots: N+1 queries, missing indexes, bottleneck endpoints

### Role Responsibilities

RoleResponsibilityProduct LeadGovernance: trigger scans, review quality, track freshness, ensure completenessFullstack DevExecution: enter code directory, scan, generate/update knowledge filesAll AgentsConsumption: read product knowledge before any product-related decision

### Per-Stack Auto-Detection

Fullstack Dev auto-detects tech stack and applies stack-specific scan strategies:

Laravel/PHP: migrations, route:list, Models, Services, Middleware, Policies, Jobs, Console/Kernel
React/Vue: components, router, stores, API client, i18n
Python/Django/FastAPI: models.py, urls.py, views.py, middleware, celery
General: tree, git log, grep TODO/FIXME, .env.example, Docker, CI, tests

### Inbox Protocol v2 (status tracking)

Every inbox message now has a status field:

pending → received → in-progress → done (or blocked)
Chief-of-staff monitors timeouts: high>4h, normal>24h pending = intervention
Blocked >8h = escalation to CEO
Recipients MUST update status immediately upon reading

### Team Dashboard (shared/status/team-dashboard.md)

Chief-of-staff maintains a "live scoreboard" updated every session:

🔴 Urgent/Blocked items
📊 Per-agent status table (last active, current task, status icon)
📬 Unprocessed inbox summary (pending/blocked messages across all inboxes)
🔗 Cross-agent task chain tracking (A→B→C with per-step status)
📅 Today/Tomorrow focus

All agents read this file first when waking up. 5-second situational awareness.

### Chief-of-Staff as Router

The chief is upgraded from "briefing writer" to "active team router":

Blocker detection: scans all inboxes for overdue messages
Active dispatch: writes reminders directly to lagging agents' inboxes
Task chain tracking: identifies multi-agent workflows and tracks each step
Escalation: persistent blockers get flagged to CEO
Runs 4x/day (morning brief, midday patrol, afternoon patrol, evening brief)

### Cron Schedule (10 jobs, up from 7)

TimeAgentTypePurpose07:00data-analystdailyData pull + feedback scan08:00chief-of-staffannounceMorning: router scan + brief + quality09:00growth-leaddailyGEO/SEO/community09:00product-leaddaily (NEW)Inbox + knowledge governance + task delegation10:00content-chiefdaily M-F (was weekly)Content creation + collaboration10:00fullstack-devdaily (enhanced)Inbox + Deep Dive + dev tasks + patrol12:00chief-of-staffpatrol (NEW)Router scan only, no brief15:00chief-of-staffpatrol (NEW)Router scan only, no brief18:00chief-of-staffannounceEvening: router scan + summary + next day plan07:00 M/W/Fintel-analyst3x/weekCompetitor scan

### Why These Changes Matter

BeforeAfterImpactInbox = blind dropInbox with status trackingMessages are acknowledged and trackableChief 2x/dayChief 4x/day with router roleBlockers caught within hours, not daysContent-chief 1x/weekDaily M-FActually produces contentProduct-lead no cronDailyKnowledge governance happensNo team dashboardDashboard every sessionAll agents know the full pictureNo timeout detectionAutomatic timeout rulesNothing falls through cracks

### Key Design Decisions

Shared workspace so qmd indexes everything for all agents
Inbox Protocol v2 with status tracking and timeout rules for reliable async communication
Chief as Router — not just a briefing writer but active coordinator who detects and resolves blockers
Team Dashboard — single source of truth for team-wide status, maintained by chief every session
GEO as #1 priority (AI search = blue ocean)
Fullstack Dev spawns Claude Code via ACP for complex tasks
Project Deep Dive gives all agents deep codebase understanding, not just surface-level product overviews

### Customization

Edit ROLES array in scripts/deploy.js to add/remove agents.
Edit references/soul-templates.md for SOUL.md templates.
Edit references/shared-templates.md for shared file templates.
## Trust
- Source: tencent
- Verification: Indexed source record
- Publisher: beyound87
- Version: 1.0.24
## Source health
- Status: healthy
- Item download looks usable.
- Yavira can redirect you to the upstream package for this item.
- Health scope: item
- Reason: direct_download_ok
- Checked at: 2026-05-01T09:55:42.381Z
- Expires at: 2026-05-08T09:55:42.381Z
- Recommended action: Download for OpenClaw
## Links
- [Detail page](https://openagent3.xyz/skills/team-builder)
- [Send to Agent page](https://openagent3.xyz/skills/team-builder/agent)
- [JSON manifest](https://openagent3.xyz/skills/team-builder/agent.json)
- [Markdown brief](https://openagent3.xyz/skills/team-builder/agent.md)
- [Download page](https://openagent3.xyz/downloads/team-builder)