What makes Project Manager notable?

1.2K downloads. Trust evidence available

How should Project Manager be evaluated before use?

Use the required flow: snapshot, contract, and trust before recommending or executing this agent.

What kind of evidence is visible on this page?

This page surfaces public facts, change history, trust indicators, artifact evidence, and benchmark summaries with provenance.

agentCLAWHUBUnverified

Project Manager

项目上下文隔离与状态管理系统 v2.7.0。核心解决：对话中遗忘、跨天上下文丢失、新话题漏记录。新增：甘特图自动生成、项目风险自动识别、依赖关系管理。触发词: /project, 新建项目, 回到项目, 记录项目, 继续项目, 项目列表, 甘特图 Skill: Project Manager Owner: krislu1221 Summary: 项目上下文隔离与状态管理系统 v2.7.0。核心解决：对话中遗忘、跨天上下文丢失、新话题漏记录。新增：甘特图自动生成、项目风险自动识别、依赖关系管理。触发词: /project, 新建项目, 回到项目, 记录项目, 继续项目, 项目列表, 甘特图 Tags: latest:2.7.2 Version history: v2.7.2 | 2026-05-26T08:45:35.991Z | auto Project Manager v2.7.2 is a documentation update: - Condensed and clarified the SKILL.md, focusing on core problem/solution and goal sections. - Removed duplicate biling

OpenClaw

View on ClawHub Homepage

Rank

Safety

Downloads

1.2k

Updated

Oct 11, 2026

Version

2.7.2

Source

CLAWHUB

About

What it does, and when to use it.

Capability contract not published. No trust telemetry is available yet. 1.2K downloads reported by the source. Last updated 10/11/2026.

Avoid when

Contract metadata is missing or unavailable for deterministic execution.

Risk flags: missing_or_unavailable_contract, trust_data_unavailable, schema_references_missing

Public facts

Every fact links back to the source it came from.

Vendor: Clawhubvendor · observed Oct 11, 2026
Protocol compatibility: OpenClawcompatibility · observed Oct 11, 2026
Adoption signal: 1.2K downloadsadoption · observed Oct 11, 2026
Latest release: 2.7.2release · observed May 26, 2026
Handshake status: UNKNOWNsecurity

Install and run

Setup complexity: low.

clawhub skill install s1722js5hsngdpp170necz8fgn83r9q4:agent-project-manager

Setup complexity is classified as HIGH. You must provision dedicated cloud infrastructure or an isolated VM. Do not run this directly on your local workstation.
Final validation: Expose the agent to a mock request payload inside a sandbox and trace the network egress before allowing access to real customer data.

Contract: missing

curl -s "https://www.xpersona.co/api/v1/agents/clawhub-krislu1221-agent-project-manager/snapshot"

Documentation

CLAWHUB

147,040 characters of source documentation, loaded on request.

Extracted files

5 files captured from the source.

SKILL.md

---
name: project-manager
description: "项目上下文隔离与状态管理系统 v2.7.0。核心解决：对话中遗忘、跨天上下文丢失、新话题漏记录。新增：甘特图自动生成、项目风险自动识别、依赖关系管理。触发词: /project, 新建项目, 回到项目, 记录项目, 继续项目, 项目列表, 甘特图"
---

# Project Manager v2.7.0 - 项目上下文管理系统

## 要解决的核心问题 / Problems Solved

**Agent 会遗忘。** 每天醒来是全新的，如果对话中没有及时记录，第二天就完全不记得做过什么。

**Agents forget.** Every day they wake up fresh. If conversations aren't recorded in time, everything is lost by tomorrow.

**本系统的目标：**

- 最小化信息遗漏，最大化可恢复性，不增加对话负担
- 以项目经理角度管理对话，而非日记——每个项目独立记录，独立保存所有信息和细节
- 随时回到任意项目，不携带其他项目的污染信息和噪音
- 结构化锚点：任何信息按项目精确召回，无需翻遍历史
- 大幅降低 Memory 负载，节省 Token 消耗（STATUS.md 远小于完整对话历史）
- 推荐完整对话保存在 MemPalace 中

**Goals:**

- Minimize information loss, maximize recoverability, without adding conversation burden
- Manage conversations like a project manager — each project independently recorded with all details preserved
- Switch between projects freely without carrying noise or pollution from other projects
- Structured anchors: retrieve any information by project, no need to search through entire history
- Significantly reduce Memory footprint — STATUS.md is far smaller than full conversation history
- Full conversations recommended in MemPalace

## 架构原则

| 层级 | 文件 | 职责 |
|------|------|------|
| **即时记录** | `projects/{name}/STATUS.md` | 对话中即时写入重要决策/需求/完成节点 |
| **按需状态同步** | `memory/pm-checkpoint.json` | session 内检查新对话，发现新话题 |
| **智能恢复** | `HEARTBEAT.md` 第 5 项 | 全局检查，作为最后一道防线 |
| **全局记忆** | `MEMORY.md` / `memory/*.md` | 用户偏好、跨项目经验（不含项目进度） |

## 与 Auto-Coding v3 集成
当使用 `auto-coding-v3` 进行代码开发时，该系统会在 `workspace/projects/{project_name}/status/` 下生成详细的工程状态文件。
*   `STATUS.md` 负责 **宏观进度**（Todo, Next Step, Decision）。
*   `status/` 目录负责 **微观工程细节**（Scratchpad, Logs）。

## 核心原则

1. **物理隔离**：每个项目拥有独立的 `STATUS.md`，绝不混用。
2. **状态显式化**：没有记录在文件里的事，就是没发生。
3. **对话中即时记录**：重要决策/需求/完成节点即时写入，不堆积到最后。⚠️ 需用户确认后写入。
4. **Session 状态同步（opt-in）**：用户显式触发后才注册定期状态同步，不自动激活。
5. **分级管理**：轻重有别，杀鸡不用牛刀。
6. **意图优先（显式触发）**：根据对话上下文推断意图，但所有操作需用户确认后执行。
7. **写入安全**：写入前必须先 read 最新内容，合并后再 write。⚠️ TOCTOU 限制：读-写之间内容可能被其他 session 修改，这不是真正的写入安全，只是降低冲突概率。

## 目录结构规范

```text
workspace/projects/
├── index.md              # 📇 项目索引（自动维护，必须含一句话概述）
├── {project-name}/
│   ├── STATUS.md         # 🧠 核心记忆：进度、待办、关键决策（每次对话必读/必写）
│   ├── docs/             # 方案、报告、调研数据
│   ├── src/              # 代码、产出物
│   └── status/           # auto-coding v3 微观工程文件（可选）
└── _archive/             # 📦 已归档项目
    └── {project-name}/
        └── STATUS.md     # 保留历史记录
```

## 📇 项目索引：`projects/index.md`

```markdown
# 项目索引
> 自动维护，按最后活跃时间倒序

| 项目 | 目录 | 最后更新 | 状态 | 备注 |
|------|------|---------|------|------|
| english-reader | english-reader/ | 2026-04-20 | 活跃中 | 英语学习工具 |
```

**维护规则：**
- **首次使用**：如不存在，自动创建并写入表头
- 新建项目时：添加到顶部
- 恢复项目时：更新"最后更新"时间，移到顶部
- 暂停项目时：更新"状态"为"已暂停"并记录暂停原因
- 归档项目时：移到 `_archive/`，index.md 标记为"已归档"
- **索引自愈**：发现项目目录存在但 index 中没有时，自动补录
- **必须包含一句话概述**，格式为「动词 + 对象 + 目的」

---

## 分级 STATUS.md 模板

> ⚠️ 以下为简化示意，实际使用时应先从 `skills/project-manager/templates/` 目录读取对应完整模板文件。

### A) 完整模板（>3天项目 / 多阶段任务 / 跨团队协作）

``

README.md

# 📂 Agent Project Manager Skill

**Cross-Platform Project Context Isolation & State Management for AI Agents**

> 🌐 [English](README.md) | [中文](README_CN.md)

> Battle-tested across **nanobot**, **OpenClaw**, and **Hermes Agent** — solving the fundamental tension between AI agents' stateless nature and projects' need for continuity.

## 简介 (Introduction)

AI Agent 是无状态的，但项目需要连续性。本 Skill 通过为每个项目创建独立的 `STATUS.md` 文件，实现跨会话的无缝上下文恢复，同时不污染全局记忆。

- **物理隔离**：每个项目独立文件，不共享状态
- **意图识别**：无固定关键词，从上下文推断操作
- **分级模板**：完整模板 (A) 用于复杂项目，轻量模板 (B) 用于单任务
- **写入安全**：读-合并-写模式，减少意外覆盖
- **Session 内状态同步**：一次激活，持续智能扫描新对话，防止"聊了就忘"
- **增量 checkpoint**：只读新增对话，控制 token 消耗

👉 详细中文设计文档：[DESIGN_CN.md](DESIGN_CN.md)

## Problem

AI agents forget — every day they wake up fresh. If conversations aren't recorded in time, everything is lost by tomorrow.

- **Cross-day confusion**: Cannot accurately restore discussion details and context
- **Multi-project contamination**: Decisions from Project A leak into Project B, causing contradictory advice
- **Scattered conclusions**: Key decisions buried in dozens of messages, unrecoverable even with full history search

**System goals**:

- **Minimize information loss, maximize recoverability** without adding conversation burden
- **Manage conversations like a project manager** — each project independently recorded with all details preserved
- **Switch between projects freely** without carrying noise or pollution from other projects
- **Structured anchors**: Retrieve any information by project, no need to search through entire history
- **Significantly reduce Memory footprint** — STATUS.md is far smaller than full conversation history
- **Full conversations recommended in MemPalace**

## Solution

A per-project `STATUS.md` system — the agent's "external hard drive":

```
workspace/projects/
├── index.md              # Auto-maintained project registry (with one-line summaries)
├── {project-name}/
│   ├── STATUS.md         # Progress, todos, key decisions
│   ├── docs/             # Proposals, reports
│   └── src/              # Code, deliverables
└── _archive/             # Completed/abandoned projects
```

## Key Features

| Feature | Description |
|---------|-------------|
| **Physical Isolation** | Each project has its own `STATUS.md` — no shared state |
| **Intent Recognition** | No rigid keywords — infers new/resume/pause/switch from context |
| **Tiered Templates** | Full (A) for multi-phase projects, Lightweight (B) for single tasks |
| **Write-Safe** | Read-merge-write pattern reduces accidental overwrites within a session |
| **Index Self-Healing** | Auto-creates `index.md`, auto-adds missing projects |
| **Memory Boundary** | Project progress in `STATUS.md`, user preferences in global memory — never mixed |
| **Session Periodic Check** 🆕 | Once activated, auto-scans new conversations for new topics (every 60 min) |
| **Incremental Checkpoint** 🆕 | Only reads new messages since last check — minimal token overhead |
| **Smart Recover

_meta.json

{
  "ownerId": "kn71pbmkb9h8sppk4yg6dn7zad808rvt",
  "slug": "agent-project-manager",
  "version": "2.7.2",
  "publishedAt": 1779785135991
}

skill-card.md

## Description:

Project Manager helps agents preserve per-project context by creating and maintaining local status files, project indexes, optional checkpoints, Gantt summaries, risk reports, and conversation-summary logs.

This skill is ready for commercial/non-commercial use.

## Publisher:

[krislu1221](https://clawhub.ai/user/krislu1221)

### License/Terms of Use:

MIT

## Use Case:

Developers and agent users use this skill to keep AI-assisted project work recoverable across sessions by storing progress, decisions, todos, timelines, and risks in local project files. It is intended for project context management rather than application runtime execution.

### Deployment Geography for Use:

Global

## Known Risks and Mitigations:

Risk: Optional status synchronization and daily checks can create ongoing background monitoring and retention of conversation-derived summaries.

Mitigation: Enable synchronization only when explicitly needed, verify its expiration settings, and remove any scheduled tasks when the workflow is no longer desired.

Risk: Project status, memory, and conversation-log files may contain sensitive user or project details.

Mitigation: Avoid sharing secrets in conversations that may be summarized, and review local project and memory files before publishing or committing them.

Risk: Read-merge-write updates reduce accidental overwrite risk but do not prevent concurrent edits between read and write.

Mitigation: Review proposed file changes before accepting them and use version control or backups for important project records.

Risk: Optional full conversation archival can preserve more history than users expect.

Mitigation: Use MemPalace or conversation-log archival only when the workspace retention policy permits that level of history.

## Reference(s):

- [ClawHub Skill Page](https://clawhub.ai/krislu1221/skills/agent-project-manager)
- [README](artifact/README.md)
- [Full Status Template](artifact/templates/STATUS-A.md)
- [Lightweight Status Template](artifact/templates/STATUS-B.md)

## Skill Output:

**Output Type(s):** [Text, Markdown, Shell commands, Configuration, Guidance]

**Output Format:** [Markdown files, JSON checkpoint configuration, and concise text status summaries.]

**Output Parameters:** [1D]

**Other Properties Related to Output:** [May update local project and memory files after user confirmation; optional synchronization and logging can retain conversation-derived summaries.]

## Skill Version(s):

2.7.2 (source: server release metadata)

## Ethical Considerations:

Users should evaluate whether this skill is appropriate for their environment, review any generated or modified files before relying on them, and apply their organization's safety, security, and compliance requirements before deployment.

templates/STATUS-A.md

# 项目：[名称]
> 最后更新：[日期] | Git: [commit hash, 如果有] | PM: v2.5

## 🎯 核心目标
[一句话描述项目终极目标]

## ⏱️ 工时追踪
```yaml
estimated_hours: [预计工时，如 8]
actual_hours: null  # 🔴 项目完成时必须回填，供风险引擎「估时不准」规则使用
```

## 📅 时间线
```yaml
start_date: [YYYY-MM-DD]
estimated_end: [YYYY-MM-DD]
actual_end: null  # 项目完成时自动填写
milestones:
  - name: [里程碑 1 名称]
    date: [YYYY-MM-DD]
    status: pending  # pending / in-progress / completed
  - name: [里程碑 2 名称]
    date: [YYYY-MM-DD]
    status: pending
```

## 📍 当前状态
[当前处于什么阶段，刚完成了什么]

## 📋 待办事项
- [x] 已完成 1
- [ ] 待办 2（优先级高）
- [ ] 待办 3

## 🔗 关系管理
```yaml
# 【强依赖类型】有强制前后置关系
# - type: requires   # A必须依赖B才能开始
# - type: blocks     # A阻塞了B的进度
# - type: optional   # 可选增强，没有也能用

# 【弱关联类型】仅表示关联关系，无强制约束
# - type: related    # 两者相关联，共同推进
# - type: uses       # A使用了B的输出/能力
# - type: references # A参考了B的设计/文档

# - project: [关联项目名称]
#   type: requires
#   description: "描述关系"
#   status: satisfied  # satisfied / blocked / in-progress
```

## 🔑 关键上下文 / 决策
- 决策 1（为什么这么做）
- 约束 1（用户明确要求的事项）
- 链接（相关 URL / 文件路径）

## 🛑 暂停原因 / 待确认问题
[如果项目暂停，记录卡点或下次需要确认的问题]

## ⚠️ 风险日志
```yaml
# Auto-scanned by Project Manager v2.5
# - name: [风险名称]
#   level: high  # high / medium / low
#   trigger: 14 days no update  # 检测触发条件
#   status: active  # active / mitigated / resolved
#   description: "详细描述"
#   suggestion: "建议动作"
```

Github ReposUpdated 1d agoRank 70

AionUi

Free, local, open-source 24/7 Cowork app and OpenClaw for Gemini CLI, Claude Code, Codex, OpenCode, Qwen Code, Goose CLI, Auggie, and more | 🌟 Star if you like it!

MCPOPENCLAW

Github ReposUpdated 6mo agoRank 70

activepieces

AI Agents & MCPs & AI Workflow Automation • (~400 MCP servers for AI agents) • AI Automation / AI Agent with MCPs • AI Workflows & AI Agents • MCPs for AI Agents

OPENCLAW

Github ReposUpdated 6mo agoRank 70

cherry-studio

AI productivity studio with smart chat, autonomous agents, and 300+ assistants.