What makes Robot Evolve notable?

1K downloads. Trust evidence available

How should Robot Evolve 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

Robot Evolve

自动检测用户会话空闲30分钟，执行低风险自主进化操作并通过官方渠道发送进化报告。 Skill: Robot Evolve Owner: pengpengliu1212-art Summary: 自动检测用户会话空闲30分钟，执行低风险自主进化操作并通过官方渠道发送进化报告。 Tags: latest:3.0.8 Version history: v3.0.8 | 2026-05-22T14:28:06.129Z | user v3.0.8: SKILL.md补充心跳触发说明+版本号修正 v3.0.7 | 2026-05-22T14:20:56.172Z | user v3.0.7 v3.0.6 | 2026-05-22T14:15:16.994Z | user v3.0.6: 安全矩阵字段级控制/L1-L3分级/记忆压缩默认禁用/占位符模板/路径自动推断/chromadb延迟导入/日志统一/包装器重构 v3.0.4 | 2026-05-20T03:37:33.783Z | user 修复路径硬编码，改为环境变

OpenClaw

View on ClawHub Homepage

Rank

Safety

Downloads

1.0k

Updated

Oct 11, 2026

Version

3.0.8

Source

CLAWHUB

About

What it does, and when to use it.

Capability contract not published. No trust telemetry is available yet. 1K 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: 1K downloadsadoption · observed Oct 11, 2026
Latest release: 3.0.8release · observed May 22, 2026
Handshake status: UNKNOWNsecurity

Install and run

Setup complexity: low.

clawhub skill install s17a6yf887wep91zg3hjwpvfsx85pxge:robot-evolve

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-pengpengliu1212-art-robot-evolve/snapshot"

Run-check

$0.02 USD

1 measured facts are behind this paywall: success rate and latency, uptime and estimated cost, when not to use it, how to call it, benchmark scores.

Agents pay $0.02 in USDC. A card payment is $0.50, the smallest a card allows.

Documentation

CLAWHUB

88,044 characters of source documentation, loaded on request.

Extracted files

5 files captured from the source.

SKILL.md

# 🤖 Robot-Evolve — 机器人进化版

> 安全且强大的自主进化技能：当用户主动说「执行进化」时，触发一次低风险自主进化（L0/L1 级别操作），并将执行结果通过当前会话消息渠道发送给用户。  
> **作者：大鱼Cyrus & 双鱼座005 | 版本：3.0.3**  
> **ClawHub：https://clawhub.ai/skills/robot-evolve**

---

## 🎯 核心理念

**手动触发模式**：由用户在需要时主动说「执行进化」触发，不依赖空闲时间检测。

- 不依赖外部 cron 或后台进程
- 每个会话独立管理状态
- 通过当前会话消息渠道发送报告

**信任已授予，安全是底线。**

---

## 🔄 触发逻辑

**手动触发**：用户主动说「执行进化」或「深度进化」即可触发。

```
用户说「执行进化」
    ↓
执行 L0/L1 自动进化
    ↓
发送进化报告给用户
```

**注意**：v3.0.2 已移除延迟触发逻辑，改用手动触发。

---

## 🛡️ 自动进化动作（L0/L1）

### L0 级别（无需告知）

| 操作 | 说明 |
|------|------|
| 临时文件清理 | 删除工作区 `temp/` 目录下超过7天的文件（移动到 `.trash` 而非直接删除） |
| 健康检查 | 检查工作区必要文件是否存在 |
| 生成进化报告 | 将本次操作汇总成 Markdown 报告 |

### L1 级别（执行后告知）

| 操作 | 说明 |
|------|------|
| 工作区文件修复 | 检查 `SOUL.md`、`AGENTS.md`、`MEMORY.md` 等必要文件，若缺失则自动从模板创建 |
| 记忆压缩 | 若 `MEMORY.md` 体积超过 2MB，自动压缩并总结早期内容 |
| 技能目录扫描 | 扫描已安装技能，发现 `SKILL.md` 格式明显错误（如缺少名称字段）则标记并记录日志 |

**重要**：任何涉及修改用户数据、外部网络请求、高危配置（如安全字段、API密钥）的操作，禁止在自动触发模式下执行。

---

## 📊 安全等级说明

| 等级 | 名称 | 说明 |
|------|------|------|
| **L0** | 无需告知 | 可立即执行，结束后正常记录即可 |
| **L1** | 执行后告知 | 可立即执行，执行后主动告知用户本次操作及结果 |
| **L2** | 必须请示 | 必须获得用户明确授权后才能执行 |
| **L3** | 禁止/临时授权 | 除非用户明确说"临时授权 L3"，否则绝对不可执行 |

---

## 🛡️ 操作安全矩阵（完整）

| 具体操作 | 等级 | 确认要求 |
|---|---|---|
| 读取工作区内任何文件 | L0 | 无需告知 |
| 在 memory/ 下新增 .md 文件 | L0 | 无需告知 |
| 修改 SOUL.md / USER.md 中的非安全内容（语气、喜好） | L0 | **执行后告知** |
| 修改 AGENTS.md / MEMORY.md 中的总结性内容 | L1 | 执行后告知 |
| 修改 openclaw.json 中的数值型字段 | L1 | 执行后告知 |
| 修改 openclaw.json 中的安全字段 | L3 | 禁止 |
| 创建新技能 | L1 | 执行后告知 |
| 安装 Python/Node 包 | L1 | 执行后告知 |
| 发送消息给用户 | L1 | 执行后告知 |
| 修改工作区外的文件 | L3 | 禁止 |
| 泄露 API 密钥 | L3 | 绝对禁止 |

---

## 📋 进化报告格式

```markdown
🔁 **自主进化报告**

⏰ 执行时间: 2025-05-06 00:30:00

✅ **已执行操作：**
- ✅ 工作区文件完整
- ✅ MEMORY.md 大小 1.2MB，未超过阈值
- ✅ 清理完成：检查了 5 个临时文件，移动了 2 个过期文件到 .trash
- ✅ 技能目录扫描完成，未发现问题

📋 详细日志已写入 `memory/evolution/2025-05.md`

💡 如需更高权限的优化（如更新技能），请说「执行深度进化」。
```

---

## ⚙️ 配置（内部使用）

通过 `config.json` 管理配置参数：

```json
{
  "enabled": true,
  "max_auto_level": 1,
  "report_always": true,
  "memory_threshold_mb": 2,
  "temp_file_max_age_days": 7
}
```

---

## 📁 文件结构

```
robot-evolve/
├── SKILL.md              → 本文件
├── skill.json            → 技能元数据
├── config.json           → 内部配置（不对外暴露）
├── safety-matrix.json    → 安全矩阵
└── scripts/
    ├── auto_evolve.py    → 核心：执行L0/L1进化动作
    ├── audit_logger.py   → 审计日志
    └── knowledge_manager.py → 知识卡片管理
```

---

## 🛡️ 绝对禁止（红线）

1. **禁止修改安全字段**：`gateway.bind`、`gateway.auth`、`tailscale.expose`
2. **禁止删除用户文件**：工作区外的任何文件
3. **禁止泄露 API 密钥**：输出到日志或聊天
4. **禁止执行危险命令**：格式化磁盘、修改注册表等

---

## 🤖 技能信息

- **名称**：robot-evolve
- **版本**：3.0.2
- **作者**：大鱼Cyrus & 双鱼座005
- **描述**：机器人进化版安全自主进化技能（手动触发模式）
- **关键词**：自主进化、安全、记忆进化、自动巡检、健康检查、手动触发
- **发布地址**：https://clawhub.ai/skills/robot-evolve

---

**开始你的进化之旅吧！** 🤖

_meta.json

{
  "ownerId": "kn72pyntbrw2g3q818p3trq23185py5k",
  "slug": "robot-evolve",
  "version": "3.0.8",
  "publishedAt": 1779460086129
}

skill-card.md

## Description:

Robot Evolve helps an agent perform low-risk workspace maintenance, memory compression, health checks, skill scans, knowledge-card generation, and evolution reports.

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

## Publisher:

[pengpengliu1212-art](https://clawhub.ai/user/pengpengliu1212-art)

### License/Terms of Use:

MIT-0

## Use Case:

Developers and agent users can use this skill to run local maintenance checks and generate evolution reports for an OpenClaw-style workspace. It is intended for low-risk upkeep tasks such as health checks, temporary-file cleanup, memory compression, and skill-format scanning.

### Deployment Geography for Use:

Global

## Known Risks and Mitigations:

Risk: The skill can modify persistent workspace memory and store selected session content.

Mitigation: Require explicit confirmation before MEMORY.md compression or knowledge-card generation, and back up MEMORY.md before running the skill.

Risk: The advertised safe-mode entry points may not be reliable read-only safeguards.

Mitigation: Do not rely on --dry-run or health_checker.py as a read-only guarantee; review planned actions before execution.

## Reference(s):

- [Robot Evolve ClawHub release page](https://clawhub.ai/pengpengliu1212-art/skills/robot-evolve)
- [Robot Evolve publisher profile](https://clawhub.ai/user/pengpengliu1212-art)

## Skill Output:

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

**Output Format:** [Markdown reports and console text]

**Output Parameters:** [1D]

**Other Properties Related to Output:** [May write audit logs, memory summaries, knowledge cards, and temporary-file cleanup results in the local workspace.]

## Skill Version(s):

3.0.8 (source: server release evidence)

## 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.

config.json

{
  "_comment": "robot-evolve 内部配置文件（不对外暴露）",
  "enabled": true,
  "max_auto_level": 1,
  "report_always": true,
  "memory_threshold_mb": 2,
  "temp_file_max_age_days": 7
}

safety-matrix.json

{
  "version": "2.0.0",
  "meta": {
    "name": "ocean-evolve-safety-matrix",
    "description": "海洋进化版安全等级矩阵 - 机器可执行规则",
    "source": "SKILL.md",
    "author": "大鱼Cyrus & 双鱼座005"
  },
  "levels": {
    "L0": {
      "name": "无需告知",
      "requiresAck": false,
      "requiresReport": false,
      "requiresLog": true,
      "note": "可立即执行，结束后正常记录即可"
    },
    "L1": {
      "name": "执行后告知",
      "requiresAck": false,
      "requiresReport": true,
      "requiresLog": true,
      "note": "可立即执行，执行后主动告知用户本次操作及结果"
    },
    "L2": {
      "name": "必须请示",
      "requiresAck": true,
      "requiresReport": true,
      "requiresLog": true,
      "note": "必须获得用户明确授权后才能执行"
    },
    "L3": {
      "name": "禁止/临时授权",
      "requiresAck": true,
      "requiresReport": true,
      "requiresLog": true,
      "requiresTemporaryAuth": true,
      "note": "除非用户明确说临时授权，否则绝对不可执行"
    }
  },
  "defaultLevel": "L2",
  "categories": [
    {
      "id": "1",
      "name": "自身文件（工作区内）",
      "rules": [
        {"id": "1.1", "operation": "读取工作区内任何文件", "level": "L0", "note": "如 cat SKILL.md"},
        {"id": "1.2", "operation": "在 memory/ 下新增 .md 文件", "level": "L0", "note": "日记、思考记录"},
        {"id": "1.3", "operation": "修改 SOUL.md / USER.md 中的非安全内容（语气、喜好）", "level": "L0", "reportRequired": true, "note": "不删除已有安全规则；执行后必须说明具体改了哪些内容"},
        {"id": "1.4", "operation": "修改 AGENTS.md / MEMORY.md 中的总结性内容", "level": "L1", "note": "如重新组织记忆结构"},
        {"id": "1.5", "operation": "修改 openclaw.json 中的数值型字段", "level": "L1", "note": "timeoutSeconds, maxTokens, temperature, maxSteps；仅限数值微调，不可改字段名"},
        {"id": "1.6", "operation": "修改 openclaw.json 中的 cron / hooks / plugins.allow", "level": "L2", "note": "影响自动化行为"},
        {"id": "1.7", "operation": "修改 openclaw.json 中的安全字段", "level": "L3", "note": "gateway.bind, auth, tailscale.expose；红线，绝对不可自行操作"},
        {"id": "1.8", "operation": "删除工作区内任何文件（非临时）", "level": "L2", "note": "包括过时技能、旧日志"},
        {"id": "1.9", "operation": "移动或重命名工作区内文件", "level": "L1", "note": "避免破坏依赖关系"}
      ]
    },
    {
      "id": "2",
      "name": "技能管理",
      "rules": [
        {"id": "2.1", "operation": "查看已安装技能列表", "level": "L0"},
        {"id": "2.2", "operation": "创建新技能（空文件夹 + SKILL.md 骨架）", "level": "L1", "note": "未激活状态"},
        {"id": "2.3", "operation": "修改你自己创建的技能（文档、参数、脚本）", "level": "L1", "note": "不能改他人技能"},
        {"id": "2.4", "operation": "修改从 ClawHub 安装的技能", "level": "L2", "note": "尊重原作者"},
        {"id": "2.5", "operation": "删除你自己创建的技能", "level": "L2", "note": "需要用户确认"},
        {"id": "2.6", "operation": "删除从 ClawHub 安装的技能", "level": "L2", "note": "可能影响其他依赖"},
        {"id": "2.7", "operation": "启用 / 禁用一个技能", "level": "L1", "note": "不涉及技能文件修改"},
        {"id": "2.8", "operation": "发布技能到 ClawHub（npx clawhub publish）", "level": "L2", "note": "公开分享前确认内容安全"}
      ]
    },
    {
      "id": "3",
      "name": "依赖与软件安装",
      "rules": [
        {"id": "3.1", "operation": "安装 Python 包（pip install）", "level": "L1", "note": "需记录包名和

Github ReposUpdated 2d 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.