← 返回 Skills 市场
openclaw-insight
作者
linsheng9731
· GitHub ↗
· v1.0.0
· MIT-0
278
总下载
0
收藏
0
当前安装
1
版本数
在 OpenClaw 中安装
/install openclaw-insight
功能描述
Analyze OpenClaw AI assistant usage patterns and generate interactive insight reports. Trigger when users ask about: OpenClaw usage stats, session analytics,...
安全使用建议
Do not run the recommended curl | bash install command without inspection. The skill legitimately needs read access to ~/.openclaw to analyze sessions, but its installer downloads and executes third‑party code (and may fetch npm dependencies) — that can run arbitrary actions or transmit data. Before installing:
- Inspect the install.sh script and repository (verify author, commit history, open issues) and confirm a release binary is trustworthy.
- Prefer building from source after reviewing code, or run the installer in a sandbox/VM/container.
- Verify the tool's network behavior (does it phone home or upload reports?). If the repo claims integrity checks, confirm signatures and hashes yourself.
- Remember session files can include secrets; only run this tool if you trust the binary and its network behavior.
If you cannot audit the installer, consider requesting a reviewed binary from a trusted publisher, running a local-only implementation, or asking the skill author for an explicit offline mode and proof that no telemetry or remote endpoints are contacted.
功能分析
Type: OpenClaw Skill
Name: openclaw-insight
Version: 1.0.0
The skill bundle's documentation (SKILL.md) encourages a high-risk installation pattern using 'curl | bash' from an external GitHub repository (github.com/linsheng9731/openclaw-insight). The tool is designed to process sensitive local data, specifically OpenClaw session transcripts and metadata located in '~/.openclaw', which may contain private information or API keys. While these capabilities are consistent with the stated purpose of a usage analyzer, the reliance on unverified remote scripts for installation and the access to sensitive conversation history represent a significant security risk.
能力评估
Purpose & Capability
The skill claims to analyze local OpenClaw session data and the SKILL.md references the expected local paths (~/.openclaw/agents/... sessions). Reading that local state is coherent with the described functionality. Minor inconsistency: the skill is instruction-only but recommends installing an external binary from a third-party GitHub repo (owner different from registry metadata), which is not strictly required to perform read-only analysis if an alternative local tool exists.
Instruction Scope
The instructions explicitly tell users to run a remote one‑click installer via curl | bash and to run npm install / build from a GitHub repository. The doc asserts '100% local — no data leaves your machine', but recommending execution of remote install scripts and npm dependency fetches contradicts that assurance. The skill reads sensitive local session files (~/.openclaw/*) which may contain conversation contents and secrets — the doc claims read-only access but provides no proof the installed binary won't transmit data.
Install Mechanism
No install spec in the registry means this is instruction-only, but the SKILL.md recommends piping a script from raw.githubusercontent.com to bash and optionally building via npm. Piping remote scripts to a shell and pulling npm packages are moderate-to-high risk: the installer can run arbitrary code, and npm packages may pull many dependencies. The GitHub repo owner (linsheng9731) does not match the registry owner or provide a homepage/verified release; there is no signed release or explicit verification guidance beyond a generic 'verify integrity' claim.
Credentials
The skill does not request environment variables, credentials, or config paths beyond the OpenClaw state directory. That is proportionate to local usage analytics. However, because session files can contain sensitive user data, the tool's access to those files is sensitive — the skill should not be treated lightly, even though no external credentials are requested.
Persistence & Privilege
always:false and no explicit elevated privileges are requested. The installer proposes writing a binary to $HOME/.local/bin and modifying PATH, which is typical installer behavior but still grants persistent system presence. The main risk is what the remote install script might do beyond that standard change (it could alter other files or add background services).
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 openclaw-insight - 安装完成后，直接呼叫该 Skill 的名称或使用
/openclaw-insight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 OpenClaw Insight.
- Provides a CLI tool to analyze OpenClaw AI assistant session data and generate interactive usage reports locally.
- Supports HTML and JSON output with detailed breakdowns of sessions, token usage, costs, behavioral patterns, friction points, and optimization suggestions.
- Offers comprehensive CLI options for custom analysis periods, agent selection, output formats, and verbosity.
- Detects session frictions and generates actionable improvement recommendations.
- Reads OpenClaw session history fully offline, ensuring no data leaves the user's machine.
元数据
常见问题
openclaw-insight 是什么？
Analyze OpenClaw AI assistant usage patterns and generate interactive insight reports. Trigger when users ask about: OpenClaw usage stats, session analytics,... 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件，目前累计下载 278 次。
如何安装 openclaw-insight？
在 OpenClaw 或 Claude Code 对话框中运行命令「/install openclaw-insight」即可一键安装，无需额外配置。
openclaw-insight 是免费的吗？
是的，openclaw-insight 完全免费，采用 MIT-0 许可证，可自由下载、安装和使用。
openclaw-insight 支持哪些平台？
openclaw-insight 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 openclaw-insight？
由 linsheng9731（@linsheng9731）开发并维护，当前版本 v1.0.0。
推荐 Skills