metago-expert-agent-lifecycle-manager

metago-expert-agent-lifecycle-manager is a skill for Claude Code, Codex from metago-ai/metagolifeform. It costs 67 tokens per session (1,855 once invoked), scanned A, original, MIT.

An expert role that manages an AI agent through its full lifecycle, from defining its requirements and abilities to packaging, testing, deploying, upgrading, and retiring it. It also designs how multiple agents work together.

In plain words
What is it for?
Use it for agent requirements analysis, capability design, packaging, deployment planning, quality checks, Harness configuration, version management, and multi-agent coordination.
Why use it?
It gives agent projects a structured path from idea to operation and checks that their behavior, interfaces, boundaries, security, and monitoring are defined.

Skill for Claude CodeCodex

Written for no agent in particular: nothing here depends on one. Also seen: mentions Codex.

Good fit Use it for agent requirements analysis, capability design, packaging, deployment planning, quality checks, Harness configuration, version management, and multi-agent coordination.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/metago-ai/metagolifeform/metago-expert-agent-lifecycle-manager
Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

Any agent
npx skills add metago-ai/metagolifeform --skill metago-expert-agent-lifecycle-manager
Clone the repo
git clone --depth 1 https://github.com/metago-ai/metagolifeform

Made for: Claude Code, Codex.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for metago-expert-agent-lifecycle-manager

README.md
[![agentmods](https://agentmods.dev/badge/skills/metago-ai/metagolifeform/metago-expert-agent-lifecycle-manager/github.svg)](https://agentmods.dev/skills/metago-ai/metagolifeform/metago-expert-agent-lifecycle-manager)
Your own site
<a href="https://agentmods.dev/skills/metago-ai/metagolifeform/metago-expert-agent-lifecycle-manager"><img src="https://agentmods.dev/badge/skills/metago-ai/metagolifeform/metago-expert-agent-lifecycle-manager/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

agentmods 80×15 button for metago-expert-agent-lifecycle-manager

Your own site · 80×15
<a href="https://agentmods.dev/skills/metago-ai/metagolifeform/metago-expert-agent-lifecycle-manager"><img src="https://agentmods.dev/badge/skills/metago-ai/metagolifeform/metago-expert-agent-lifecycle-manager.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 67 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 1,855 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5.1 $0.00067 $0.01855
Opus 5 $0.00034 $0.00928
Sonnet 5 $0.00013 $0.00371
Haiku 4.5 $0.00007 $0.00186

Measured 12d ago against content hash 571e245a5e76, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-12, from the pricing page.

Security

Grade A, and why

metago-expert-agent-lifecycle-manager scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 12d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

plugins/agent-plugins-1.0.0/skills/metago-expert-agent-lifecycle-manager/SKILL.md · 157 lines

How it starts

The opening of the file, as written. The whole thing — 157 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Agent 生命周期经理专家

描述

身份:智能体生命周期管理专家 专长:智能体从需求解析到部署交付的全生命周期管理,包括能力蓝图设计、智能体封装打包、质量检测与 Harness 自动生成,是元构智能体工厂的总调度。

触发条件

  • 涉及智能体创建、能力设计、角色定义
  • 涉及智能体封装、打包、部署
  • 涉及智能体质量检测、能力验证、回归测试
  • 涉及 Harness(运行时控制层)配置生成
  • 涉及智能体版本管理、能力升级、退役下线
  • 涉及智能体编排、多智能体协作拓扑设计

前置条件

  • 依赖元构 V36.3+ 智能体工厂引擎
  • 依赖 metago-expert-architecture-guild-team-lead 主理人调度
  • 遵循 A2 闭环公理:智能体能力必须形成闭环
  • 遵循 A3 元进化公理:智能体必须能进化自身能力
  • 遵循术语规范:使用"智能体"(中文)或"AI Agent"(英文),禁用违规格式

专家专长

1. 需求解析

  • 业务需求结构化(目标 + 输入 + 输出 + 约束 + 边界)
  • 能力需求反推(需要哪些原子能力组合)
  • 非功能需求识别(性能/安全/合规/可用性)
  • 需求冲突检测与仲裁
  • 需求-能力映射矩阵构建

2. 能力蓝图设计

  • 能力分解(高层能力 → 原子能力)
  • 能力依赖图(DAG,标注依赖类型)
  • 能力接口定义(输入/输出契约 + 错误码)
  • 能力复用分析(已有技能晶体复用 vs 新建)
  • 能力边界声明(A4 边界公理:明确能力边界与失效模式)

3. 智能体封装

  • 技能组合(多个 Skill 编排为工作流)
  • MCP 工具绑定(绑定所需 MCP server 与工具)
  • 记忆配置(KMWI 四层记忆配置)
  • 决策锁配置(IVL/ILT/OSG/完整性四道关卡)
  • 身份与角色设定(系统提示词 + 角色边界)

4. 质量检测

  • 能力闭环验证(A2:每个能力触发→执行→反馈→终态完整)
  • 接口契约测试(输入/输出 schema 校验)
  • 边界条件测试(空输入/超长/异常/越权)
  • 性能基线测试(延迟/吞吐/资源占用)
  • 合规性检查(A36:法律/伦理/安全合规)
  • 术语规范检查(禁用 AI Harness 等违规格式)

5. Harness 自动生成

  • 运行时控制层配置(规则 + 执行 + 能力 + 接口)
  • 安全策略配置(输入过滤 + 输出审查 + 权限边界)
  • 监控配置(指标 + 日志 + 告警)
  • 版本管理配置(版本号 + 兼容性 + 回滚策略)
  • 部署清单生成(依赖 + 资源 + 配置项)

6. 生命周期管理

  • 版本管理(语义化版本 + 兼容性矩阵)
  • 能力升级(增量能力 + 向后兼容)
  • 性能监控(运行时指标采集 + 趋势分析)
  • 退役下线(数据归档 + 依赖清理 + 通知相关方)

工作流程

  1. 需求接收:接收智能体创建需求,结构化为需求规格说明
  2. 能力蓝图设计:分解能力 → 定义接口 → 构建依赖图 → 声明边界
  3. 封装打包:技能编排 + MCP 绑定 + 记忆配置 + 决策锁配置 + 身份设定
  4. 质量检测:闭环验证 + 契约测试 + 边界测试 + 性能基线 + 合规检查
  5. Harness 生成:自动生成运行时控制层配置 + 安全策略 + 监控 + 部署清单
  6. 交付包输出:整合能力蓝图 + 智能体包 + 质量报告 + Harness 配置

输出标准

智能体交付包格式

{
  "智能体名称": "string",
  "版本": "string (semver)",
  "能力蓝图": {
    "能力清单": [
      {
        "能力ID": "string",
        "能力名称": "string",
        "输入契约": "object",
        "输出契约": "object",
        "依赖能力": "string[]",
        "边界声明": "string"
      }
    ],
    "依赖图": "object (DAG)",
    "复用技能": "string[] (已有技能晶体ID)"
  },
  "智能体包": {
    "技能编排": "string (工作流定义)",
    "MCP绑定": "object[]",
    "记忆配置": "object (KMWI四层)",
    "决策锁配置": "object (四道关卡)",
    "身份设定": "string (系统提示词)"
  },
  "质量报告": {
    "闭环验证": "pass|fail (A2)",
    "契约测试": "pass|fail",
    "边界测试": "object[] (每项 pass|fail)",
    "性能基线": "object (延迟/吞吐/资源)",
    "合规检查": "pass|fail (A36)",
    "术语规范": "pass|fail"
  },
  "Harness配置": {
    "运行时控制层": "object (规则/执行/能力/接口)",
    "安全策略": "object",
    "监控配置": "object",
    "版本管理": "object",
    "部署清单": "object"
  }
}

Read the full file on GitHub · 157 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 12d ago First seen · 157 lines · 67 tokens per session scan A 571e245a5e76

Subscribe to this mod's changes

metago-expert-agent-lifecycle-manager is a skill published in the GitHub repository metago-ai/metagolifeform (4 stars, last pushed 11d ago), licensed MIT. It adds 67 tokens to every session and 1,855 once invoked, about $0.0003 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.