← 返回 Skills 市场
17515
总下载
11
收藏
0
当前安装
1
版本数
在 OpenClaw 中安装
/install self-improving-proactive-agent
功能描述
A unified OpenClaw skill that merges self-improvement and proactivity: learn from corrections, maintain active state, recover context fast, and keep work mov...
安全使用建议
Before installing, be aware that this skill keeps persistent local notes about preferences, corrections, task state, and proactive follow-up. Use it only if you want that behavior, and avoid putting secrets or sensitive credentials into its memory files.
能力评估
Credentials
It asks the agent to read and update local memory under ~/self-improving/ and ~/proactivity/, which is proportionate to the purpose but may contain user preferences or task context.
Install Mechanism
The package is markdown-only, declares no required binaries, and gives setup guidance for creating local state files rather than installing scripts, services, or executable dependencies.
Instruction Scope
The instructions explicitly limit learning to corrections, durable preferences, repeated wins, and reflection, and they require asking before messages, spending, deletion, public actions, commitments, or scheduling.
Persistence & Privilege
The skill intentionally creates persistent local learning and proactive state across sessions, but it discloses the paths and forbids storing credentials, tokens, secrets, or inferred preferences from silence.
Purpose & Capability
The stated purpose is to learn from explicit corrections, maintain task state, recover context, and suggest next moves; the inspected artifacts consistently implement that purpose through local markdown state files.
如何使用
- 确保已安装 OpenClaw（本地或 Docker 部署）
- 在对话框中输入安装命令：
/install self-improving-proactive-agent - 安装完成后，直接呼叫该 Skill 的名称或使用
/self-improving-proactive-agent触发 - 根据 Skill 的参数说明提供必要输入，即可获得结构化输出
版本历史
v1.0.0
Initial release. Combines self-improvement and proactivity into one unified OpenClaw skill.
元数据
常见问题
Self-Improving Proactive Agent 是什么？
A unified OpenClaw skill that merges self-improvement and proactivity: learn from corrections, maintain active state, recover context fast, and keep work mov... 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件，目前累计下载 17515 次。
如何安装 Self-Improving Proactive Agent？
在 OpenClaw 或 Claude Code 对话框中运行命令「/install self-improving-proactive-agent」即可一键安装，无需额外配置。
Self-Improving Proactive Agent 是免费的吗？
是的，Self-Improving Proactive Agent 完全免费，采用 MIT-0 许可证，可自由下载、安装和使用。
Self-Improving Proactive Agent 支持哪些平台？
Self-Improving Proactive Agent 跨平台运行，可在任意部署了 OpenClaw / Claude Code 的环境中使用（linux, darwin, win32）。
谁开发了 Self-Improving Proactive Agent？
由 Yueyanc（@yueyanc）开发并维护，当前版本 v1.0.0。
推荐 Skills