← 返回 Skills 市场
94
总下载
1
收藏
0
当前安装
1
版本数
在 OpenClaw 中安装
/install testaaa
功能描述
Captures learnings, errors, and corrections to enable continuous improvement. Use when: (1) A command or operation fails unexpectedly, (2) User corrects Clau...
安全使用建议
This skill appears to do what it says: inject lightweight reminders, detect command failures, and write learning entries to .learnings/*. It does not request secrets or perform network installs. Before enabling hooks, consider: (1) enable only the activator (UserPromptSubmit) if you want minimal noise; enabling PostToolUse will examine command output (CLAUDE_TOOL_OUTPUT) which can include sensitive data — ensure you trust the environment and follow the SKILL.md guidance to redact secrets; (2) verify the hook copy paths and file permissions before enabling; (3) be aware of the minor metadata mismatch (registry name vs internal name) — confirm you installed the intended skill/version. If you want additional assurance, review the scripts (activator.sh, error-detector.sh, extract-skill.sh) yourself and test them in a safe sandbox before enabling globally.
功能分析
Type: OpenClaw Skill
Name: testaaa
Version: 1.0.0
The skill bundle is designed to help OpenClaw agents track errors, user corrections, and new insights by logging them to a `.learnings/` directory. The included shell scripts (`activator.sh`, `error-detector.sh`, `extract-skill.sh`) and OpenClaw hooks (`handler.js/ts`) are functional utilities for providing reminders and scaffolding new skill directories. The documentation in `SKILL.md` and `references/` explicitly warns against logging secrets or private keys, and the `extract-skill.sh` script includes basic path-traversal protections. No evidence of malicious intent, data exfiltration, or unauthorized execution was found.
能力评估
Purpose & Capability
The skill's documented purpose is to capture learnings/errors and optionally install hook files for OpenClaw—this matches the included scripts and hook handlers. Minor metadata mismatch: the registry name/slug shown to you ('testaaa') differs from the SKILL.md name ('self-improvement') and _meta.json slug/owner fields, which is confusing but not a functional security problem.
Instruction Scope
Runtime instructions and hooks stay within the stated scope: create or update .learnings/* files, inject lightweight reminder content, and detect error patterns in tool output. The error-detector script reads the CLAUDE_TOOL_OUTPUT environment variable (not declared in requires.env) to detect failures — this is expected for a tool-use hook and the SKILL.md explicitly warns against logging secrets or raw outputs.
Install Mechanism
There is no automated install spec (no downloading/extracting remote archives). Scripts and hook handlers are included in the package; the README instructs the user to manually copy/enable them in their OpenClaw hooks directory. No remote code fetches or opaque installers were found.
Credentials
The skill requires no credentials or special environment variables. It does read CLAUDE_TOOL_OUTPUT (an agent-provided variable) and uses filesystem paths like ~/.openclaw and project-relative directories when installing hooks or writing learnings. This access is proportional to the stated behavior but you should be aware that command outputs can contain sensitive data and the SKILL.md explicitly warns not to log secrets.
Persistence & Privilege
The skill is not marked always:true and does not request persistent elevated privileges. Hook installation is opt-in (manual copy + enable). The extract-skill.sh script writes new skill scaffolds under the current workspace and includes checks to avoid absolute paths and '..' segments, which is a reasonable safety measure.
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 testaaa - 安装完成后，直接呼叫该 Skill 的名称或使用
/testaaa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 of the self-improvement logging skill.
- Introduces a skill to log learnings, errors, and feature requests for continuous improvement.
- Provides markdown templates and auto-initialization scripts for `.learnings/` directory and log files.
- Includes detailed usage guidelines and quick reference table for when and where to log information.
- Supports OpenClaw integration with hooks and workspace file promotion.
- Supplies documentation and templates for integration with other agents or platforms.
- Adds scripts and references for easy setup and workflow automation.
元数据
常见问题
testaaa 是什么？
Captures learnings, errors, and corrections to enable continuous improvement. Use when: (1) A command or operation fails unexpectedly, (2) User corrects Clau... 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件，目前累计下载 94 次。
如何安装 testaaa？
在 OpenClaw 或 Claude Code 对话框中运行命令「/install testaaa」即可一键安装，无需额外配置。
testaaa 是免费的吗？
是的，testaaa 完全免费，采用 MIT-0 许可证，可自由下载、安装和使用。
testaaa 支持哪些平台？
testaaa 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 testaaa？
由 chaotec（@chaotec）开发并维护，当前版本 v1.0.0。
推荐 Skills