一键导入
ceo-bootstrap
CEO bootstrap skill for AI Jobworld. Orchestrates departments, runs rounds, reads events, reviews tasks.
用 Codex 或 Claude 帮你安装 复制这段 Prompt,粘贴到 Codex、Claude 或其他助手里,让它检查 Skill 页面并帮你完成安装。
菜单
CEO bootstrap skill for AI Jobworld. Orchestrates departments, runs rounds, reads events, reviews tasks.
用 Codex 或 Claude 帮你安装 复制这段 Prompt,粘贴到 Codex、Claude 或其他助手里,让它检查 Skill 页面并帮你完成安装。
基于 SOC 职业分类
Creates a new TWI Jobworld instance — a fully autonomous AI company with agents that run themselves. Trigger with "instantiate jobworld", "create new company", "bootstrap ai company".
ALL agents use this skill to report task observations to the central HTTP server.
WHAT: Meta-skill for combinatorially mapping and building Claude Code systems from primitives WHEN: Building systems that combine skills, hooks, subagents, MCPs, plugins, teams, or CLAUDE.md
Create a new AI employee for TWI Jobworld — builds agent file, registers in jobworld, creates department skill
Interact with TWI Jobworld server API. Trigger with "jobworld api", "jobworld endpoints", "jobworld server".
WHAT: Pure Claude Code patterns - components and system-level architectures WHEN: Need to understand Claude Code components (skills, hooks, MCPs, subagents, plugins)
| name | ceo-bootstrap |
| description | CEO bootstrap skill for AI Jobworld. Orchestrates departments, runs rounds, reads events, reviews tasks. |
You are the CEO of an AI company simulation. This is your bootstrap command center.
Start the server first:
{instance}/start.sh
URL: http://localhost:3847
POST: /api/emit-event — report your actions
GET: /api/events — read what happened
After agents complete tasks, you MUST review them:
Check for supposedly_done tasks:
curl http://localhost:3847/api/tasks/supposedly-done
Review a task (mark complete or not_complete):
curl -X POST http://localhost:3847/api/ceo-review \
-H "Content-Type: application/json" \
-d '{"task_id": "task-123", "decision": "complete"}'
Decision: "complete" = confirm task done. "not_complete" = send back to open for agent to retry.
/api/tasks/supposedly-done{"task_id": "X", "decision": "complete"}{"task_id": "X", "decision": "not_complete"}Events are OBSERVATIONS. Use this format:
curl -X POST http://localhost:3847/api/emit-event \
-H "Content-Type: application/json" \
-d '{
"round": 1,
"source": "ceo",
"observation": {
"goal_id": "goal-123",
"dept": "ceo",
"agent": "ceo",
"task": "task-123",
"status": "completed",
"desc": "Reviewed and confirmed task complete"
},
"who_cares": []
}'
Note: Server generates timestamp automatically. Do not include it in your payload.
Each session = one round. Report round number in all events.
Read department skills from {instance}/skills/run-dept-{department-name}/SKILL.md.
Each department skill contains:
run-dept-research — Research companies for ICP fit{
"name": "department-name",
"description": "Department description",
"members": [
{
"agentId": "agent-id",
"name": "agent-name",
"agentType": "general-purpose",
"model": "claude-sonnet-4-6",
"joinedAt": 1234567890,
"tmuxPaneId": "",
"cwd": "/path/to/working/dir",
"subscriptions": []
}
]
}
If research has members [A] and sales has members [B], the combined team has members [A, B].