A smart utility to monitor, visualize, and manage your AI session's token consumption and remaining context window.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install context-usage-checker
2. Manual Installation
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
3. Prompt Installation
Copy this prompt to OpenClaw to install it automatically.
Help me install context-usage-checker using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Prefer to download?
Get the raw skill files in a ZIP archive.
What is Context Usage Checker?
The Context Usage Checker is a specialized tool designed for the Openclaw Skills ecosystem to provide real-time transparency into an AI's memory. By tracking token consumption within a conversation, it helps users avoid the sudden performance degradation or 'forgetfulness' that occurs when an AI model hits its context limit.
This skill transforms raw data into a human-readable dashboard, featuring visual progress bars and intelligent status alerts. It allows developers and power users to maintain long-running sessions by knowing exactly when to compress context or start fresh conversations, ensuring a seamless experience across all Openclaw Skills integrated platforms.
Context Usage Checker Use Cases
- Monitoring remaining capacity during long coding sessions or complex research tasks.
- Receiving proactive alerts when session context reaches critical levels (70% and 90% thresholds).
- Optimizing API costs and performance by tracking token density across different models.
- Deciding when to use commands like /compact or /new based on visual usage data.
How Context Usage Checker Works
- The user triggers the skill via natural language queries or specific slash commands like /context.
- The skill analyzes the current session history and calculates the character count for both Chinese and English text.
- It applies a specialized estimation algorithm (approx. 1.5 chars/token for Chinese, 4 chars/token for English) to determine total usage.
- The system identifies the specific model's context window (e.g., 256k for qwen3.5-plus).
- It generates a visual report including a 20-character progress bar, percentage breakdown, and status-based recommendations.
Context Usage Checker Setup
To integrate this utility into your Openclaw Skills environment, ensure your agent has access to the repository and follow these steps:
# Navigate to your skills directory
cd path/to/openclaw/skills
# Clone the skills repository if you haven't already
git clone https://github.com/rfdiosuao/openclaw-skills
# The Context Usage Checker is enabled by default once the skill file is present.
You can verify the installation by typing /usage in your chat interface.
Context Usage Checker Data Schema & Taxonomy
The skill organizes usage data based on the model's technical specifications and current session metadata. Below is the primary data structure used for the report:
| Attribute | Description | Logic/Default |
|---|---|---|
| Model ID | The active LLM being used | Auto-identified (e.g., qwen-max) |
| Token Limit | Maximum capacity of the model | 32k to 256k depending on model |
| Warning Threshold | Trigger for yellow alert | 70% usage |
| Critical Threshold | Trigger for red alert | 90% usage |
| Estimation Error | Accuracy margin | ±10% vs official platform counts |
Context Usage Checker Advanced Features
- Auto-Model Recognition: Automatically adjusts context limits for Qwen series and falls back to a 128k default for unknown models.
- Smart Threshold Alerts: Dynamically suggests corrective actions like /new or /compact when limits are nearing exhaustion.
- Customizable Constants: Developers can modify WARNING_THRESHOLD and CRITICAL_THRESHOLD directly in the skill source code.
- Cross-Platform Compatibility: Works natively across Feishu, Telegram, WhatsApp, and other Openclaw Skills supported interfaces.
- High Performance: Optimized for low-latency environments with query execution times under 10ms.
SKILL.md
Loading
Related Openclaw Skills
Smart PDF OCR
An intelligent document processing skill powered by the MinerU API for high-precision OCR and layout analysis of scanned PDFs.
Code Explore
A specialized skill for inspecting and mapping the structure, logic, and data flow of local codebases without modifying files.
Code-Write
A precision-focused skill for AI agents to implement or modify code in local projects while strictly adhering to existing project styles and handling edge cases.
Code Architect
Code Architect generates minimal, actionable implementation plans and architectural designs optimized for your existing local codebase.
AI PDF Converter
A high-performance AI document processing tool that converts PDFs into structured formats like Markdown, HTML, and DOCX using intelligent layout analysis.
Excel Export
A specialized utility for AI agents to generate professional, multi-sheet Excel workbooks from structured JSON data using Openclaw Skills.
Featured*
