← 返回 Skills 市场
17404
总下载
43
收藏
190
当前安装
1
版本数
在 OpenClaw 中安装
/install self-reflection
功能描述
Continuous self-improvement through structured reflection and memory
安全使用建议
Install only if you want an agent to keep local self-review memory over time. Do not log secrets, private user data, tokens, or sensitive project details, and inspect the external GitHub CLI before symlinking or running it because that executable was not included in this reviewed package.
功能分析
Type: OpenClaw Skill
Name: self-reflection
Version: 1.1.1
The OpenClaw AgentSkills skill bundle 'self-reflection' is classified as benign. The skill's purpose is to enable continuous self-improvement for an AI agent by tracking mistakes and logging lessons. It integrates with OpenClaw's heartbeat mechanism, instructing the agent via `SKILL.md` to run local `self-reflection` commands (`check`, `read`, `log`) and store data in local markdown and JSON files (`~/workspace/memory/self-review.md`, `~/.openclaw/self-review-state.json`). There is no evidence of data exfiltration, malicious execution, persistence mechanisms, or prompt injection with a harmful objective. The required binaries (`jq`, `date`) are standard utilities, and the installation steps are typical for a CLI tool.
能力评估
Purpose & Capability
The stated purpose is continuous self-improvement through logged lessons, and the documented commands, heartbeat checks, memory file, and state file fit that purpose.
Instruction Scope
It asks users to add heartbeat instructions that periodically run reflection checks and may trigger read/log actions; this is disclosed and user-enabled, but it changes ongoing agent behavior.
Install Mechanism
The reviewed artifact contains markdown and JSON only; the README instructs cloning a GitHub repository and symlinking its bin/self-reflection executable, so users should inspect that external CLI before running it.
Credentials
The documented data flow is local reflection text and timer state in user-controlled paths, with no shown credential access, account access, network exfiltration, or elevated privileges.
Persistence & Privilege
The skill intentionally persists reflection history and timer state across sessions; this is core to the feature and disclosed, but logged content may influence future agent behavior.
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-reflection - 安装完成后，直接呼叫该 Skill 的名称或使用
/self-reflection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
v1.1.1
Initial ClawHub release: Continuous self-improvement through structured reflection
元数据
常见问题
Self Reflection 是什么？
Continuous self-improvement through structured reflection and memory. 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件，目前累计下载 17404 次。
如何安装 Self Reflection？
在 OpenClaw 或 Claude Code 对话框中运行命令「/install self-reflection」即可一键安装，无需额外配置。
Self Reflection 是免费的吗？
是的，Self Reflection 完全免费（开源免费），可自由下载、安装和使用。
Self Reflection 支持哪些平台？
Self Reflection 跨平台运行，可在任意部署了 OpenClaw / Claude Code 的环境中使用（cross-platform）。
谁开发了 Self Reflection？
由 hopyky（@hopyky）开发并维护，当前版本 v1.1.1。
推荐 Skills