🔧

Model Usage

Name: Model Usage
Author: openclaw

🌐Community

by openclaw · v2026.3.2 · MITRepository

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.

Install on your platform

We auto-selected Claude Code based on this skill’s supported platforms.

Run in terminal (recommended)

terminal

claude mcp add model-usage npx -- -y @trustedskills/model-usage

Or manually add to ~/.claude/settings.json

~/.claude/settings.json

{
  "mcpServers": {
    "model-usage": {
      "command": "npx",
      "args": [
        "-y",
        "@trustedskills/model-usage"
      ]
    }
  }
}

Requires Claude Code (claude CLI). Run claude --version to verify your install.

About This Skill

What it does

OpenClaw is a personal AI assistant designed to be run locally on user devices. It provides responses across various messaging platforms like WhatsApp, Telegram, and Slack, and offers voice interaction capabilities on macOS/iOS/Android. A key feature includes the ability to render a live, controllable Canvas interface for visual interaction.

When to use it

Personal Productivity: Use OpenClaw as an always-on assistant to manage tasks, set reminders, or quickly access information without relying on cloud services.
Cross-Platform Communication: Receive and respond to messages across multiple platforms (WhatsApp, Telegram, Slack, Discord) from a single interface.
Local Data Privacy: Benefit from having your AI interactions processed locally, enhancing privacy and reducing reliance on external servers.
Customizable Interface: Utilize the Canvas feature for creating custom visual workflows or dashboards within the assistant environment.

Key capabilities

Cross-platform messaging support (WhatsApp, Telegram, Slack, Discord, etc.)
Voice interaction on macOS/iOS/Android devices
Local execution for enhanced privacy
Live, controllable Canvas interface
Personalized AI assistance

Example prompts

"OpenClaw, remind me to call John at 2 PM."
"Show me the latest news headlines on my Canvas."
"Send a message to the team in Slack saying 'Meeting is postponed.'"
“What’s the weather like today?”

Tips & gotchas

The preferred setup method involves running openclaw onboard in your terminal, which guides you through initial configuration.
OpenClaw's functionality relies on local processing; performance may be affected by device resources.

View Repository →

TrustedSkills Verification

Unlike other registries that point to live repositories, TrustedSkills pins every skill to a verified commit hash. This protects you from malicious updates — what you install today is exactly what was reviewed and verified.

Verified Commit7b5e64ef →

Installing this skill downloads the exact code at commit 7b5e64ef, not the current state of the repository. This prevents supply-chain attacks from unauthorized updates.

Security Audits

Gen Agent Trust Hub	Pass
Socket	Pass
Snyk	Pass

Details

Version: v2026.3.2
License: MIT
Author: openclaw
Installs: 0
Updated: Mar 3, 2026
Published: Nov 24, 2025

Repository (canonical source) →

🌐 Community

Passed automated security scans.

Pinned commit7b5e64ef

Install command fetches the verified snapshot, not the live repository.