Model Usage
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. Trigger when asked for model-level usage/cost data from codexbar, or when you need a scriptable per-model summary from codexbar cost JSON.
v1.0.0
2026/01/06
- Initial release of the model-usage skill for summarizing per-model usage and cost from CodexBar CLI local cost logs. - Supports both Codex and Claude providers. - Allows fetching summaries for the current (most recent) model or all models. - Provides both text and JSON output formats. - Can process costs directly from CodexBar CLI output or from a JSON file/stdin. - Includes guidance for CLI use and script automation.
Security Scan
状态
clean
OpenClaw
gpt-5.5
OpenClaw 分析
This skill appears to do what it says: install and run a local usage-cost CLI, then summarize the results.
置信度: high
VirusTotal
Type: OpenClaw Skill Name: model-usage Version: 1.0.0 The skill's purpose is to summarize local cost data from the `codexbar` CLI tool. The `SKILL.md` instructs the agent to install `codexbar` via Homebrew cask, which is a standard package manager. The `scripts/model_usage.py` script executes `codexbar cost` using `subprocess.check_output` to retrieve local usage data, then parses and formats it. The script does not perform any network calls, access sensitive files (e.g., `~/.ssh`, `~/.aws`), or attempt to exfiltrate data. The `references/codexbar-cli.md` confirms that `codexbar cost` reads local logs. There are no prompt injection attempts in the markdown files that would lead to malicious actions. All operations are aligned with the stated purpose and lack high-risk behaviors or malicious intent.