The component provides live character count, word count, approximate token count, remaining token capacity, usage percentage, warning state, and limit-exceeded state. It supports Detailed, Compact, and Inline display modes along with configurable token limits, warning thresholds, whitespace normalization, optional metrics, and a progress bar.

The reusable OnEstimateChanged event allows consuming applications to monitor prompt usage and disable AI submission when the configured limit is exceeded.

Token values are estimated using approximately one token per four characters. Actual token usage varies by AI model, tokenizer, language, formatting, source code, and special characters. A provider-specific tokenizer should be used when exact token counts are required.