项目经理 Agent

SKILL效率

原名 project-manager-agent

效率 接触敏感数据 许可证明确 OpenClaw
安装这个技能

project-manager-agent 用于补充效率相关能力,适合在 OpenClaw 中需要让 Agent 承接效率相关任务时使用。可结合来源仓库、安装命令和原始 README 继续核验具体用法。安装前建议确认权限范围、维护状态,以及是否会触发联网、命令执行或文件读写。

累计安装 9,949 本周 423
GitHub 星标 未标注 仓库 project-manager-agent
下载量 3,486 技能市场下载数
权限级别 接触敏感数据 安装前请确认风险范围
许可证 许可证明确 MIT-0

安装

ClawHub 市场
在支持 Skills 的客户端里执行,装完问一句「你有哪些技能」验证
openclaw skills install project-manager-agent

技能档案

效率
字段来自技能仓库与公开索引,中文说明由本站标注
作者 / 维护方austindixson仓库 project-manager-agent · 版本 0.1.0
宿主客户端OpenClaw技能可被这些客户端加载
安装方式ClawHub 市场安装包 austindixson/project-manager-agent
权限级别接触敏感数据安装前请确认风险范围
许可证MIT-0许可证明确,商用前请自行确认条款
仓库可用性available技能仓库当前是否可访问
最近核验2026-05-01上游最后核验时间 · 数据更新 未标注
正文体量4.6 KBSKILL.md 原文大小,越大说明说明书写得越细

能做什么

4 项能力
上游给出的能力概览与适用场景
  • 统一调用多种 LLM
  • 支持 Claude、Gemini、Kimi 等模型
  • 适合聊天、代码和推理任务
  • 可作为 Agent 模型调用入口
调用多模型代码和文本生成Agent 推理流程OpenRouter 模型接入

权限与风险

接触敏感数据
装之前先读完这段

安装与使用指南

安装步骤

本站只展示命令,不托管安装包,也不自动执行。

怎么用

安装后应在对应宿主中按原始 README 的触发条件使用;当前草稿只做目录说明,具体调用方式由规则抽取,异常条目再复核。

SKILL.md 全文

4.6 KB
模型实际读到的说明书原文,来自技能仓库

Project Manager Agent

Description

Monitors active sub-agents, detects stalls, provides status updates, and steers stuck agents via sessions_send.

Project Manager Agent

This skill provides oversight for your OpenClaw sub-agents. It proactively checks the status of running sub-agents, identifies potential stalls or failures, and can steer stalled sub-agents by sending a continue message via the sessions_send tool.

Usage

Running the monitor

python3 <skill-dir>/scripts/project_manager.py

Options

  • --staleness_threshold_minutes <int>: Minutes after which a sub-agent is considered stalled (default from config or 10).
  • --check_all_sessions: Check all sessions, not just subagents of the current main session.
  • --json: Print human summary then a JSON line with stalledSessionKeys, steerMessage, and summary for agent steering.
  • --no_cooldown: Output all stalled keys (ignore 15-minute cooldown).
  • --record-steered KEY [KEY ...]: Record these session keys as just steered (for cooldown). Run after calling sessions_send.

Example (Status check, human-only)

python3 /Users/ghost/.openclaw/workspace/skills/project-manager-agent/scripts/project_manager.py --staleness_threshold_minutes 15

Example (Agent run with steering)

python3 /Users/ghost/.openclaw/workspace/skills/project-manager-agent/scripts/project_manager.py --json --staleness_threshold_minutes 10

Then parse the last line (JSON), call sessions_send for each key in stalledSessionKeys, and optionally run --record-steered <keys>.

Features

  • Sub-agent Monitoring: Lists all active sub-agents (file-based; no gateway API in the script).
  • Stall Detection: Flags sub-agents that haven't updated their status within a configurable time period.
  • Failure Identification: Reports on sub-agents that have aborted their last run.
  • Status Reports: Human-readable summaries and optional JSON with stalledSessionKeys for steering.
  • Automated Steering: When you run the skill with --json, parse the output and call sessions_send for each stalled session (see Agent instructions below).

Agent instructions (mandatory when running this skill)

When the user or a cron asks you to run the project-manager-agent skill:

  1. Run the script with --json (and existing args as needed):
    bash python3 /Users/ghost/.openclaw/workspace/skills/project-manager-agent/scripts/project_manager.py --json --staleness_threshold_minutes 10
  2. Parse the script stdout: The last line is a single JSON object with stalledSessionKeys, steerMessage, and summary. Parse it (e.g. take the last line that starts with { and ends with }).
  3. If stalledSessionKeys is non-empty:
    - For each session key in stalledSessionKeys (cap at 5 per run to avoid spam), call sessions_send(sessionKey, steerMessage from JSON or "Please continue working on your task.", 0) (fire-and-forget; timeoutSeconds=0).
    - Optionally run the script again to record cooldown so the same session is not steered again for 15 minutes:
    bash python3 /Users/ghost/.openclaw/workspace/skills/project-manager-agent/scripts/project_manager.py --record-steered <key1> <key2> ...
  4. Reply to the user/cron with the status summary (the human-readable part of the script output) and, if you steered any, add: "Steered N stalled subagent(s)."

You have access to sessions_list and sessions_send; the script only produces the list of session keys to steer. Steering is done by you in the same turn via the sessions_send tool.

Config (optional)

See config.json for: staleness_threshold_minutes, steer_cooldown_minutes, steer_message, max_steer_per_run.

Integration as a Cron Job

This skill is ideal for a cron job, running periodically (e.g., every 5 minutes). The agent runs the script with --json, steers stalled subagents via sessions_send, and announces the status.

Example Cron Job Payload:

{
  "payload": {
    "kind": "agentTurn",
    "message": "Run project-manager-agent skill and report status of all sub-agents.",
    "model": "openrouter/google/gemini-2.5-flash",
    "thinking": "low",
    "timeoutSeconds": 60
  },
  "schedule": {
    "kind": "every",
    "everyMs": 300000
  },
  "delivery": {
    "mode": "announce"
  },
  "sessionTarget": "isolated",
  "name": "Project Manager (Sub-agent Monitor)"
}

version: 0.2.0

安全信息

审计结论
未标注不代表已通过审计
安全审计VirusTotal、ClawScan、Static analysis第三方安全审计平台给出的结论

热门专题

按场景挑技能
全部专题 →

技能内容以作者仓库为准;分类、权限级别与许可证状态的中文说明为中文整理,最近核验 2026-05-01。