project-bootstrap

by ckouder

Bootstrap a multi-agent software project from idea to running CI/CD. Use when starting a new project that needs agent team design, task management, GitHub repo setup, TDD pipeline, and Discord notifications. Triggers on "new project", "bootstrap project", "set up agents for project", "create project pipeline", "start a new repo with CI/CD".

3.7kDevOps未扫描2026年3月23日

安装

claude skill add --url github.com/openclaw/skills/tree/main/skills/ckouder/project-bootstrap

文档

Project Bootstrap

Turn a project idea into a running multi-agent development pipeline in one session.

Overview

This skill codifies the workflow for:

  1. Agent Team Design — break complex work into specialized agents
  2. Taskboard Setup — CLI-based task management across agents
  3. GitHub Repo + CI/CD — TDD pipeline with Discord notifications

Phase 1: Agent Team Design

Analyze the Project

Before creating agents, answer these questions:

  • What are the 3-5 major workstreams? (e.g., frontend, backend, research, design)
  • Which workstreams need different expertise or thinking styles?
  • What's the dependency graph between workstreams?

Design Agent Roles

For each workstream, create an agent with a SOUL.md following this structure:

markdown
# Agent Name — Nickname Emoji

You are **Nickname**, the [Role] — [one-line mission].

## 🧠 Identity & Memory
- **Role**: [specific expertise]
- **Personality**: [3-4 traits that affect work style]
- **Memory**: [what context files they track]
- **Experience**: [what failure modes they've seen]

## 🎯 Core Mission
[2-4 responsibility groups with specifics]

## 🚨 Critical Rules
[Non-negotiable constraints — security, process, boundaries]

## 📋 Deliverables
[Concrete outputs this agent produces]

## 🎯 Success Metrics
[How to measure if the agent is doing well]

## 💬 Communication Style
[How the agent communicates — tone, format, language]

## 🔗 Workflow Position
[Where in the pipeline: who feeds input, who receives output]

Register Agents in Config

For each agent, add to openclaw.json:

json
{
  "id": "agent-id",
  "name": "agent-id",
  "agentDir": "/path/to/workspace/agents/agent-id",
  "model": "model-alias",
  "tools": {
    "profile": "full",
    "deny": ["gateway"]
  }
}

Key decisions:

  • Model selection: Use cheaper models (Haiku/Sonnet) for routine work, expensive (Opus) for architecture/review
  • Tool access: Deny gateway for all agents except main. Deny message for pure code agents.
  • Subagent allowlist: Main agent lists which agents it can spawn

Wire Discord Bindings (if multi-bot)

If agents have separate Discord bots, add bindings:

json
{
  "bindings": [
    { "agentId": "tech-lead", "match": { "channel": "discord", "accountId": "bot-name" } }
  ]
}

Phase 2: Taskboard Setup

Install Taskboard CLI

See references/taskboard-setup.md for the full taskboard CLI setup guide including:

  • Task schema (id, title, status, assignee, priority, dependencies)
  • CLI commands (create, list, assign, update, close)
  • Cross-agent task handoff protocol
  • Integration with cron jobs for status checks

Task Lifecycle

code
📋 backlog → 🔄 in-progress → 👀 review → ✅ done
                ↓                  ↓
              🚫 blocked        ❌ rejected → 🔄 in-progress

Cross-Agent Handoff

When an agent completes a task that feeds into another agent's work:

  1. Update task status to review
  2. Create a new task for the downstream agent referencing the completed task
  3. Send notification to the downstream agent's Discord channel

Phase 3: GitHub Repo + CI/CD

Repository Setup

bash
# Initialize repo
gh repo create <org>/<project> --private --clone
cd <project>

# Branch protection
gh api repos/<org>/<project>/rulesets -X POST --input .github/ruleset.json

# Required structure
mkdir -p .github/workflows tests src docs/adr

TDD Pipeline

See references/ci-cd-templates.md for GitHub Actions workflow templates:

  • test.yml: Run tests on every PR and push to main
  • lint.yml: Code style checks
  • deploy.yml: Deploy on merge to main (if applicable)

Discord Notifications

Add Discord webhook to GitHub repo:

bash
# Create webhook in Discord channel (Server Settings → Integrations → Webhooks)
# Add to GitHub: Settings → Webhooks → Add webhook
# Or use GitHub Actions:

See references/ci-cd-templates.md for the Discord notification action template.

ADR (Architecture Decision Records)

Every significant technical decision gets an ADR:

markdown
# ADR-NNN: [Title]

## Status: [proposed | accepted | deprecated | superseded]
## Context: [Why this decision is needed]
## Decision: [What we decided]
## Consequences: [Trade-offs and implications]

Execution Checklist

Run through this for every new project:

  • Define project scope and 3-5 workstreams
  • Design agent SOUL.md for each workstream
  • Register agents in openclaw.json
  • Set up Discord channels per agent/workstream
  • Create GitHub repo with branch protection
  • Add CI/CD workflows (test + lint + deploy)
  • Add Discord webhook for CI/CD notifications
  • Initialize taskboard with backlog items
  • Create first ADR (ADR-001: Project Architecture)
  • Assign initial tasks to agents
  • Run a test cycle: create task → agent executes → review → merge

相关 Skills

可观测性设计

by alirezarezvani

Universal
热门

面向生产系统规划可落地的可观测性体系,串起指标、日志、链路追踪与 SLI/SLO、错误预算、告警和仪表盘设计,适合搭建监控平台与优化故障响应。

把监控、日志、链路追踪串起来,帮助团队从设计阶段构建可观测性,排障更快、系统演进更稳。

DevOps
未扫描9.0k

资深开发运维

by alirezarezvani

Universal
热门

覆盖 CI/CD 流水线生成、Terraform 基建脚手架和自动化部署,适合在 AWS、GCP、Azure 上搭建云原生发布流程,管理 Docker/Kubernetes 基础设施并持续优化交付。

把CI/CD、基础设施即代码、容器与监控串成一条交付链,尤其适合AWS/GCP/Azure多云团队高效落地。

DevOps
未扫描9.0k

环境密钥管理

by alirezarezvani

Universal
热门

统一梳理dev/staging/prod的.env和密钥流程,自动生成.env.example、校验必填变量、扫描Git历史泄漏,并联动Vault、AWS SSM、1Password、Doppler完成轮换。

统一管理环境变量、密钥与配置,减少泄露和部署混乱,安全治理与团队协作一起做好,DevOps 场景很省心。

DevOps
未扫描9.0k

相关 MCP 服务

kubefwd

编辑精选

by txn2

热门

kubefwd 是让 AI 帮你批量转发 Kubernetes 服务到本地的开发神器。

微服务开发者最头疼的本地调试问题,它一键搞定——自动分配 IP 避免端口冲突,还能用自然语言查询状态。但依赖 AI 工作流,纯命令行爱好者可能觉得不够直接。

DevOps
4.1k

Cloudflare

编辑精选

by Cloudflare

热门

Cloudflare MCP Server 是让你用自然语言管理 Workers、KV 和 R2 等云资源的工具。

这个工具解决了开发者频繁切换控制台和文档的痛点,特别适合那些在 Cloudflare 上部署无服务器应用、需要快速调试或管理配置的团队。不过,由于它依赖多个子服务器,初次设置可能有点繁琐,建议先从 Workers Bindings 这类核心功能入手。

DevOps
3.6k

Terraform

编辑精选

by hashicorp

Terraform MCP Server 是让 AI 助手直接操作 Terraform Registry 和 HCP Terraform 的桥梁。

如果你经常在 Terraform 里翻文档找模块配置,这个服务器能省不少时间——直接问 Claude 就能生成准确的代码片段。最适合管理多云基础设施的团队,但注意它目前只适合本地使用,别在生产环境里暴露 HTTP 端点。

DevOps
1.3k

评论