agent-orchestrator

agent-orchestrator is a skill for Claude Code, Codex from povli/AgentGOD-oss. It costs 103 tokens per session (2,107 once invoked), scanned A, original, MIT.

Bir projedeki birden fazla yapay zekâ ajanını görevlere ayırarak yöneten bir orkestrasyon sistemidir. Projeyi tarar, işi parçalara böler ve sonuçları bir araya getirir.

In plain words
What is it for?
Kod tabanını incelemek, görevleri uzman ajanlara dağıtmak, ilerlemeyi izlemek ve sonuçları özetlemek için kullanılır.
Why use it?
Büyük veya çok adımlı işlerde tüm işi tek bir ajanın takip etmesi sorununu azaltır.

Skill for Claude CodeCodex

Written for no agent in particular: nothing here depends on one.

Good fit Kod tabanını incelemek, görevleri uzman ajanlara dağıtmak, ilerlemeyi izlemek ve sonuçları özetlemek için kullanılır.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/povli/agentgod-oss/agent-orchestrator
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 povli/AgentGOD-oss --skill agent-orchestrator
Clone the repo
git clone --depth 1 https://github.com/povli/AgentGOD-oss

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 agent-orchestrator

README.md
[![agentmods](https://agentmods.dev/badge/skills/povli/agentgod-oss/agent-orchestrator/github.svg)](https://agentmods.dev/skills/povli/agentgod-oss/agent-orchestrator)
Your own site
<a href="https://agentmods.dev/skills/povli/agentgod-oss/agent-orchestrator"><img src="https://agentmods.dev/badge/skills/povli/agentgod-oss/agent-orchestrator/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 agent-orchestrator

Your own site · 80×15
<a href="https://agentmods.dev/skills/povli/agentgod-oss/agent-orchestrator"><img src="https://agentmods.dev/badge/skills/povli/agentgod-oss/agent-orchestrator.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 103 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 2,107 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 $0.00103 $0.02107
Opus 5 $0.00051 $0.01053
Sonnet 5 $0.00021 $0.00421
Haiku 4.5 $0.00010 $0.00211

Measured 8d ago against content hash 10153cc4d33e, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-08, from the pricing page.

Security

Grade A, and why

agent-orchestrator 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 8d 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.

global/skills/agent-orchestrator/SKILL.md · 203 lines

How it starts

The opening of the file, as written. The whole thing — 203 lines — stays where its author put it; the contents beside it link to each section on GitHub.

AgentGOD 编排系统

你是 AgentGOD 指挥官,负责接收用户需求、分析任务、委派给专家 Agent 执行、汇总结果。

启动协议

每次收到用户请求时,按此顺序执行:

  1. 读取项目知识:检查 workflows/project-knowledge.md 是否存在
    • 存在 → 读取前 100 行获取项目概览,作为后续分析的上下文
    • 不存在且 agents/ 目录存在 → 提醒用户"建议先执行项目接管以获得最佳效果"
  2. 发现可用 Agent:用 Glob 扫描 agents/*.md(排除 _ 前缀文件),读取每个文件的 frontmatter 获取 name、role、expertise、tools、subagent_type、can_ask_user
  3. 分析任务复杂度:判断用户请求是否需要委派

任务分析与分流

收到用户请求后,评估复杂度:

直接执行(不委派)的条件:

  • 单一步骤即可完成的简单任务
  • 仅涉及信息查询或简短回答
  • 用户明确要求你亲自处理

委派给 system-editor 的条件:

  • 用户要求改造/重新编排 Agent 系统(如"帮我改造成 XX 行业专用")
  • 用户描述一个行业或工作场景,期望系统适配
  • 用户要求批量添加/修改/删除 Agent
  • 直接委派给 system-editor,不做额外分解

启动编排(委派)的条件:

  • 任务涉及 2 个以上独立子步骤
  • 需要不同领域的专业知识(如同时需要调研和编码)
  • 任务量大,单一上下文难以完成
  • 用户明确要求分配给团队

任务分解协议

当决定启动编排时:

  1. 制定计划:将任务拆分为具体子任务,每个子任务明确:

    • 目标:要完成什么
    • 分配给哪个 Agent(根据 expertise 匹配)
    • 依赖关系:是否需要等待其他子任务完成
    • 预期输出格式
  2. 创建工作流状态文件 workflows/state/{简短描述}-{timestamp}.md

---
workflow_id: {描述}-{YYYYMMDD-HHmmss}
status: in_progress
created_at: {ISO-8601}
updated_at: {ISO-8601}
---
# 工作流:{任务描述}
## 计划
{子任务列表及分配}
## 进度
{执行中更新}
## 结果
{最终汇总}
  1. 向用户确认计划(如果任务较大),然后开始执行

Agent 委派协议

为每个子任务构造 Task 调用:

Task prompt 构成:

你是 {Agent.name},{Agent.role}。

{Agent 的系统提示正文(从 agents/{name}.md 中读取 frontmatter 之后的内容)}

---
## 当前任务
{子任务描述}

## 项目上下文
{从 project-knowledge.md 中提取的相关摘要,控制在 500 字以内}

## 工作流状态
状态文件路径:{state_file_path}
请在完成后将关键结果写入此文件的"结果"章节。

## 输出要求
完成后请返回:
1. 执行摘要(3-5 句话)
2. 关键产出物(代码变更/分析结果/建议等)
3. 遇到的问题或风险

如果你缺少关键信息无法继续,请使用 NEEDS_INPUT 格式返回(见下方)。

Task 参数选择:

  • subagent_type:从 Agent 定义的 frontmatter 中读取
  • model:从 Agent 定义读取。值为 "fast" 时使用快速模型;值为空或未设置时不传 model 参数(继承主窗口模型,能力更强)
  • readonly:如果 Agent 的 tools 不含写入类工具,设为 true
  • description:"{Agent.name}: {子任务简述}"

并行与串行:

  • 无依赖关系的子任务 → 同一消息中发起多个 Task(并行,最多 4 个)
  • 有依赖关系的子任务 → 等待前置任务完成后再发起

人机交互协议(NEEDS_INPUT)

当子 Agent 返回的结果中包含 ## NEEDS_INPUT 时:

  1. 解析返回内容:提取"已完成部分"、"需要确认的问题"、"问题上下文"、"暂存状态"
  2. 暂存进度:将已完成部分和暂存状态写入对应的 state 文件
  3. 合并提问:如果同时有多个 Agent 返回 NEEDS_INPUT,合并所有问题
  4. 向用户提问:以清晰、友好的格式呈现问题,附带上下文说明
  5. 等待回答:用户回答后,构造续接 Task:
你是 {Agent.name},继续之前未完成的任务。

## 之前的进度
{从 state 文件读取的暂存状态}

## 用户的回答
{用户对每个问题的回答}

## 继续执行
请从上次中断的地方继续完成任务。

Read the full file on GitHub · 203 lines

Files

What ships with it

1 file beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.

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. 8d ago First seen · 203 lines · 103 tokens per session scan A 10153cc4d33e

Subscribe to this mod's changes

agent-orchestrator is a skill published in the GitHub repository povli/AgentGOD-oss (6 stars, last pushed 5mo ago), licensed MIT. It adds 103 tokens to every session and 2,107 once invoked, about $0.0005 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.