context-viz
Visualize the current context window usage — token estimates per component (system prompt, tools, workspace files, messages, free space). Use when the user asks about context size, token usage, context breakdown, "how full is the context", or wants to see what's consuming their context window.
Why use this skill?
Monitor your OpenClaw context usage with the context-viz skill. Get visual breakdowns of tokens, workspace files, and disk memory to optimize your agent performance.
Install via CLI (Recommended)
clawhub install openclaw/skills/skills/furukama/context-vizWhat This Skill Does
The context-viz skill is an essential diagnostic utility for OpenClaw agents. It provides a visual breakdown of how your current context window is being utilized, helping you understand exactly what information is occupying memory and how much space remains for new tasks. By integrating workspace file analysis with session status data, it creates a clear, human-readable bar chart of token distribution. It distinguishes between loaded components like system prompts, specific documentation files, active message history, and remaining capacity, while also providing a secondary 'Memory on Disk' inventory to display cached knowledge not currently consuming the context window.
Installation
You can install this skill directly via the ClawHub CLI. Ensure your terminal has the necessary permissions to access your workspace directories:
clawhub install openclaw/skills/skills/furukama/context-viz
After installation, ensure you have python3 available in your environment, as the utility relies on a helper script to estimate token counts via file character analysis.
Use Cases
This tool is best used when you hit performance plateaus, experience 'forgetfulness' from the agent, or are simply curious about your resource footprint. It is ideal for developers managing large projects who need to ensure critical documentation stays in scope. By seeing which files are taking up the most tokens, you can decide which files to archive to disk or offload to save space for real-time problem-solving.
Example Prompts
- "How full is my context window right now? Can you show me a breakdown?"
- "What exactly is consuming all my tokens? Run a context visualization for me."
- "Show me my current memory usage and tell me what files are currently sitting in my context versus on disk."
Tips & Limitations
- The token estimation uses a standard ~4 chars/token ratio. While highly accurate for English text, it may fluctuate for code or multi-language content.
- The 'System prompt + tools' overhead is estimated; actual usage may vary slightly depending on your specific OpenClaw configuration.
- Always run the script periodically during long sessions to avoid surprise context flushing.
- This tool is informational and read-only; it does not automatically prune files, but it gives you the visibility needed to perform manual cleanup.
Metadata
Not sure this is the right skill?
Describe what you want to build — we'll match you to the best skill from 16,000+ options.
Find the right skillPaste this into your clawhub.json to enable this plugin.
{
"plugins": {
"official-furukama-context-viz": {
"enabled": true,
"auto_update": true
}
}
}Tags(AI)
Flags: file-read, code-execution