← 返回 Skills 市场
Model Usage
作者
JustAskNudge
· GitHub ↗
· v0.1.0
572
总下载
0
收藏
3
当前安装
1
版本数
在 OpenClaw 中安装
/install model-usage-skill
功能描述
Use CodexBar CLI local cost usage to summarize per-model usage for Codex or Claude, including the current (most recent) model or a full model breakdown. Trig...
安全使用建议
This skill appears to do what it says: run the CodexBar CLI or parse a CodexBar JSON export and summarize per-model costs. Before installing, confirm you trust the Homebrew tap (steipete/tap) used to install codexbar, since third-party taps have higher supply-chain risk than official packages. Also note that CodexBar itself will read local cost logs under your home directory (as documented) when you run it — if you don't want that, run the bundled Python script on an exported JSON file instead. If you'd like greater assurance, review the upstream codexbar source or the installed cask before installing.
功能分析
Type: OpenClaw Skill
Name: model-usage-skill
Version: 0.1.0
The skill is designed to summarize AI model usage costs from the local CodexBar CLI. The `SKILL.md` provides clear instructions for the agent to run the bundled Python script or `codexbar` itself, without any prompt injection attempts. The `scripts/model_usage.py` script executes the `codexbar` binary using `subprocess.check_output` with a list of arguments, which is a safe method that prevents shell injection. Input parameters are validated via `argparse` choices. The script processes JSON output from `codexbar` or a specified file/stdin, extracting data without executing arbitrary code. There is no evidence of data exfiltration, persistence, unauthorized network activity, or obfuscation. All observed behavior aligns with the stated purpose.
能力评估
Purpose & Capability
Name/description ask for per-model cost summaries from CodexBar; the skill only requires the codexbar CLI and a Python script that calls 'codexbar cost' or reads a provided JSON file. Required binaries and OS restriction (darwin) match the stated purpose.
Instruction Scope
SKILL.md and the script stay on-topic: they run 'codexbar cost --format json --provider', accept file/stdin input, parse local cost JSON, and render summaries. The instructions do not direct the agent to read unrelated system files or network endpoints.
Install Mechanism
Install uses a Homebrew cask from the steipete/tap (third-party) tap. This is expected to install the codexbar binary but carries the usual trust considerations for third-party brew taps (moderate supply-chain risk compared to official packages).
Credentials
The skill declares no environment variables, no credentials, and the code does not attempt to read undisclosed env vars. Access to local cost logs is performed indirectly via the codexbar CLI or user-supplied JSON, which is appropriate for the task.
Persistence & Privilege
Skill is not always-enabled and does not request persistent system-wide privileges. It does not modify other skills or global agent config.
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 model-usage-skill - 安装完成后，直接呼叫该 Skill 的名称或使用
/model-usage-skill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
v0.1.0
- First release of the model-usage skill.
- Summarizes per-model usage and cost from CodexBar CLI logs for Codex or Claude.
- Supports both “current model” (latest daily entry) and full model breakdowns.
- Accepts input from CodexBar CLI output, file, or stdin.
- Provides flexible output formats: text or JSON.
- macOS support with homebrew install instructions; Linux guidance coming soon.
元数据
常见问题
Model Usage 是什么？
Use CodexBar CLI local cost usage to summarize per-model usage for Codex or Claude, including the current (most recent) model or a full model breakdown. Trig... 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件，目前累计下载 572 次。
如何安装 Model Usage？
在 OpenClaw 或 Claude Code 对话框中运行命令「/install model-usage-skill」即可一键安装，无需额外配置。
Model Usage 是免费的吗？
是的，Model Usage 完全免费（开源免费），可自由下载、安装和使用。
Model Usage 支持哪些平台？
Model Usage 跨平台运行，可在任意部署了 OpenClaw / Claude Code 的环境中使用（darwin）。
谁开发了 Model Usage？
由 JustAskNudge（@justasknudge）开发并维护，当前版本 v0.1.0。
推荐 Skills