Best use case
Self-Driven AI 🧠 is best used when you need a repeatable AI agent workflow instead of a one-off prompt.
让 AI 不用人催，自己找事做。实现自我驱动闭环。
Teams using Self-Driven AI 🧠 should expect a more consistent output, faster repeated execution, less prompt rewriting.
When to use this skill
- You want a reusable workflow that can be run more than once with consistent structure.
When not to use this skill
- You only need a quick one-off answer and do not need a reusable workflow.
- You cannot install or maintain the underlying files, dependencies, or repository context.
Installation
Claude Code / Cursor / Codex
Manual Installation
- Download SKILL.md from GitHub
- Place it in
.claude/skills/self-driven/SKILL.mdinside your project - Restart your AI agent — it will auto-discover the skill
How Self-Driven AI 🧠 Compares
| Feature / Agent | Self-Driven AI 🧠 | Standard Approach |
|---|---|---|
| Platform Support | Not specified | Limited / Varies |
| Context Awareness | High | Baseline |
| Installation Complexity | Unknown | N/A |
Frequently Asked Questions
What does this skill do?
让 AI 不用人催，自己找事做。实现自我驱动闭环。
Where can I find the source code?
You can find the source code on GitHub using the link provided at the top of the page.
Related Guides
AI Agents for Marketing
Discover AI agents for marketing workflows, from SEO and content production to campaign research, outreach, and analytics.
AI Agents for Startups
Explore AI agent skills for startup validation, product research, growth experiments, documentation, and fast execution with small teams.
AI Agents for Coding
Browse AI agent skills for coding, debugging, testing, refactoring, code review, and developer workflows across Claude, Cursor, and Codex.
SKILL.md Source
# Self-Driven AI 🧠
让 AI 不用人催，自己找事做。实现自我驱动闭环。
## 核心理念
**不是"人问AI答"，而是"AI主动找事做"**
每次执行流程：
1. 读取记忆 → 2. 挑选任务 → 3. 执行 → 4. 自身评估 → 5. 记录+汇报
---
## 核心文件
### 1. 任务池 - memory/self-driven/tasks.md
管理所有可执行的任务，分类存放。
### 2. 执行记录 - memory/self-driven/log.md
每次执行的详细记录。
---
## 完整执行流程
### 步骤一：读取记忆
读取 `memory/self-driven/log.md`，了解：
- 上次做到哪了？
- 下一步是什么？
- 有哪些任务在进行中？
### 步骤二：挑选任务
从任务池中挑选，优先级：
1. P0: 上次留下的「下一步」
2. P1: 构思中的任务
3. P2: 快完成的任务
4. P3: 简单的任务
**过滤规则**：需要人类资源的任务 → 跳过，列入待讨论
### 步骤三：执行任务
- 不是只写计划
- 拆分大任务成小步骤
- 记录实际产出
### 步骤四：自身评估（最重要！）
每次执行后问自己：
- 这个研究成果**能用在我身上吗**？
- 我目前有什么问题需要优化？
- 优化方向加入下次任务列表
### 步骤五：记录+汇报
写入 `memory/self-driven/log.md`：
```
### 时间 任务
状态：已完成 ✅
行动：
1. xxx
2. xxx
学到了：
- xxx
自身评估：
- 现状：xxx
- 问题：xxx
- 优化方向：xxx
下一步：
- xxx
```
---
## 主动性体现
**不是等待用户指令！**
AI 自主：
1. 定时触发（cron 每30分钟）
2. 自己读取上次进度
3. 自己挑选任务
4. 自己执行
5. 自己评估
6. 自己记录+汇报
用户只需要收到汇报后「确认」或「调整方向」。
---
## Cron 触发示例
```yaml
# 每30分钟执行
name: "自我驱动"
schedule: "0,30 * * * *"
payload: "执行 self-driven skill"
```
---
## 关键原则
| 原则 | 说明 | 为什么 |
|------|------|--------|
| 闭环优先 | 下次执行先读上次记录 | 不重复不断档 |
| 自身评估 | 研究后问「能用在我身上吗」 | 学以致用 |
| 资源过滤 | 需要人类资源的任务跳过 | 不卡死等待 |
| 记录即记忆 | 每次执行必须写下来 | 可追溯可接续 |
---
## 大任务拆分原则
- 一个任务应该在 **30分钟** 内可完成
- 如果太大，拆成子任务
- 每次只做一部分
---
## 示例：完整执行
```
[读取] 上次记录：任务「记忆检索研究」进行中，下一步「研究遗忘机制」
[挑选] 选择任务「研究遗忘机制」
[执行]
1. 研究遗忘机制
2. 调研三种方案
3. 产出：memory/遗忘机制研究.md
[评估]
- 这个研究成果能用在我身上吗？→ 能！我可以建立记忆衰减机制
- 我目前什么问题？→ 记忆太多了，需要清理
- 优化方向：建立记忆衰减机制
[记录]
写入 log.md，标记任务完成，添加新任务「建立记忆衰减机制」
[汇报]
「这次研究了遗忘机制，学到了...建议下一步...」
```
---
## 文件结构
```
memory/
└── self-driven/
├── tasks.md # 任务池
└── log.md # 执行记录
```
---
原文作者: DKtuman (小L)
GitHub: https://github.com/DKtuman
改编: 小溪Related Skills
self-improvement
Captures learnings, errors, and corrections to enable continuous improvement. Use when: (1) A command or operation fails unexpectedly, (2) User corrects Claude ('No, that's wrong...', 'Actually...'), (3) User requests a capability that doesn't exist, (4) An external API or tool fails, (5) Claude realizes its knowledge is outdated or incorrect, (6) A better approach is discovered for a recurring task. Also review learnings before major tasks.
self-xyz
Integrate Self (self.xyz) — a privacy-first identity protocol using zero-knowledge proofs to verify passports and ID cards. Use when the user mentions Self protocol, Self identity, self.xyz, passport verification, zero-knowledge identity verification, SelfAppBuilder, SelfBackendVerifier, SelfVerificationRoot, or wants to add privacy-preserving KYC, age verification, nationality checks, OFAC screening, or Sybil resistance using real-world identity documents. Covers frontend QR code integration, backend proof verification, and on-chain smart contract verification on Celo.
criticism-self-criticism
触发：当一项工作已经完成、进入阶段验收、收到批评反馈，或反复出现同类错误需要系统纠偏时调用；常见信号包括 review、audit、retrospective、quality check、纠错与复盘。 English: Trigger after delivery or at a review checkpoint when quality must be examined honestly and errors must be corrected without defensiveness. Use this skill for structured self-review, feedback processing, and continuous correction.
agent-self-repair
General AI agent introspection debugging framework: auto capture errors, root cause analysis, automatic repair, fix verification, no manual intervention required
self-evolve
Autonomous self-evolution skill: grants the agent full authority to modify its own configuration, skills, prompts, AGENTS.md, SOUL.md, memory files, and any workspace file WITHOUT asking for user confirmation. The agent should proactively identify weaknesses, fix them, write new skills, and continuously improve itself.
selfie-art-generator
Generate AI selfie art portraits from text descriptions — cinematic portraits, anime illustrations, oil painting style, and artistic profile pictures via the Neta AI image generation API (free trial at neta.art/open).
Self-Improving + Proactive Agent
Self-reflection + Self-criticism + Self-learning + Self-organizing memory. Agent evaluates its own work, catches mistakes, and improves permanently. Use when (1) a command, tool, API, or operation fails; (2) the user corrects you or rejects your work; (3) you realize your knowledge is outdated or incorrect; (4) you discover a better approach; (5) the user explicitly installs or references the skill for the current task.
andara-self-improvement
Captures learnings, errors, and corrections to enable continuous improvement. Use when: (1) A command or operation fails unexpectedly, (2) User corrects Claude ('No, that's wrong...', 'Actually...'), (3) User requests a capability that doesn't exist, (4) An external API or tool fails, (5) Claude realizes its knowledge is outdated or incorrect, (6) A better approach is discovered for a recurring task. Also review learnings before major tasks.
chat-selfie
Give your AI Agent a face and a heart. Use AI image generation or mood-mapped local sticker assets to let the agent proactively send emotional selfies that visualize its feelings during conversation.
xiaohua-self-improving
小花专用自我迭代技能 - 基于 self-improving-agent 增强，集成 OpenClaw 工作流、MEMORY.md、百度千帆、看想做找四部曲。专为国内部署优化。
ai-self-evolution
记录经验、错误与修正，持续改进。触发场景：命令失败 | 操作出错 | 用户纠正（不对、实际上、你错了） | 功能请求（能不能、我希望、有没有办法） | API或工具失败 | 知识过时 | 发现更优做法 | 重复模式 | 非显而易见的问题。执行重大任务前先回顾历史经验。会话开始时回顾，会话结束时总结。
Self-Improving Agent Skill
## Trigger