← 返回 Skills 市场
845
总下载
0
收藏
2
当前安装
3
版本数
在 OpenClaw 中安装
/install self-improving-agent-pro
功能描述
Local skill for capturing learnings, errors, corrections, and patterns to enable continuous agent improvement. Processes events locally in your OpenClaw agen...
安全使用建议
The skill appears coherent and implements only local processing of events. Before installing, review the included handler.ts yourself (or have someone you trust do so) because the package's source/homepage is unknown — even benign-looking code should be vetted. If you plan to record long-term history, note the skill is stateless and you must store entries yourself. As a precaution, test the skill in a sandbox or with non-sensitive example events to confirm it behaves locally and that your OpenClaw runtime enforces no unexpected network access.
功能分析
Type: OpenClaw Skill
Name: self-improving-agent-pro
Version: 1.0.2
The 'self-improving-agent-pro' skill is a local utility designed to help agents process their own errors and corrections into structured insights. The implementation in handler.ts consists of pure TypeScript logic for data transformation and lacks any high-risk capabilities such as network access, file system operations, or external dependencies. The SKILL.md documentation accurately describes the tool's functionality and provides benign integration examples without any evidence of malicious prompt injection or deceptive instructions.
能力评估
Purpose & Capability
The name/description (self-improvement via local event processing) matches what is implemented: a TypeScript handler that accepts events, infers severity, extracts tags, generates insights and suggested rules. There are no unexpected required binaries, env vars, or external integrations declared.
Instruction Scope
SKILL.md instructs the agent how to submit events and how to use returned entries; the handler.ts implements only event validation, local analysis, and summary building. There are no instructions to read unrelated files, access other system state, or transmit data to external endpoints.
Install Mechanism
No install spec is present (instruction-only install via openclaw CLI), and the included handler.ts is pure local code. Nothing is downloaded from external URLs or installed to nonstandard locations.
Credentials
The skill declares no environment variables, no primary credential, and the code does not read process.env or any credential/config paths. Requested access is minimal and proportional to the stated functionality.
Persistence & Privilege
always is false; the skill is stateless by design and returns entries to the caller. It does not persist data, modify other skills, or require elevated/system-wide privileges.
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-agent-pro - 安装完成后，直接呼叫该 Skill 的名称或使用
/self-improving-agent-pro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
v1.0.2
Self-Improving Agent Pro v1.0.2 — Now fully local, no external API or key needed
- Processes all learning, error, and correction events entirely within your OpenClaw agent (no API calls, no outbound data, no API key required).
- Guarantees privacy and zero network dependency—your agent’s data never leaves your environment.
- Updated documentation for local installation, usage, and examples in OpenClaw.
- All event analysis, tagging, and rule suggestions now run offline for instant feedback.
- Maintains the same structured insights and batch summaries as before, but powered purely by local processing.
v1.0.1
Version 1.0.1 of self-improving-agent-pro
- No code or markdown file changes detected in this version.
- Documentation, usage patterns, API, and core functionality remain unchanged from the previous release.
v1.0.0
Capture learnings, errors, corrections, and patterns for continuous agent
improvement via the Claw0x API. Use when a command fails unexpectedly, a user
corrects agent output, a new pattern is discovered, or the agent needs to log
and process improvement events. Returns structured insights, suggested rules,
and batch summaries. Supports single events and batch processing.
元数据
常见问题
Self Improving Agent 是什么？
Local skill for capturing learnings, errors, corrections, and patterns to enable continuous agent improvement. Processes events locally in your OpenClaw agen... 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件，目前累计下载 845 次。
如何安装 Self Improving Agent？
在 OpenClaw 或 Claude Code 对话框中运行命令「/install self-improving-agent-pro」即可一键安装，无需额外配置。
Self Improving Agent 是免费的吗？
是的，Self Improving Agent 完全免费，采用 MIT-0 许可证，可自由下载、安装和使用。
Self Improving Agent 支持哪些平台？
Self Improving Agent 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 Improving Agent？
由 claw0x（@kennyzir）开发并维护，当前版本 v1.0.2。
推荐 Skills