AI Agent Hub
Back to plugins
🧠

dsh-token-usage-ledger

Memory Updated 2026.08.25

Run the following command in DeepSeek Harness:

dsh plugin install xie-tj/dsh-token-usage-ledger

Paste the following prompt into your AI chat to install this plugin:

You can install this plugin by running `dsh plugin install xie-tj/dsh-token-usage-ledger`, with the full source code address at https://github.com/xie-tj/dsh-token-usage-ledger.

About this plugin

When using DeepSeek Harness, tracking token usage is essential for cost control and performance optimization, but the default Web profile may lack persistent recording and intuitive analysis tools, leading to data loss or difficult monitoring. The dsh-token-usage-ledger plugin addresses this by providing a persistent usage ledger that automatically records input, output, and cache tokens for each provider dispatch, with session history backfill ensuring historical data completeness.

Core capabilities include independent persistence, precise statistics, and a visual dashboard. Users can view usage data filtered by provider, model, and time range in Settings → Usage, with charts supporting daily granularity, line and bar graphs, and separate display of cache hits. Even for failed or retried requests, the plugin accurately records data, ensuring reliability, while using compact units like K/M/B/T for large numbers to simplify readability.

This plugin is ideal for developers, data analysts, or anyone using the DeepSeek Harness Web profile who needs to monitor token usage patterns. It helps users optimize resource allocation, track API call efficiency, and simplify usage analysis through an intuitive interface, providing deep insights without complex configuration to enhance overall user experience and cost management.

Use Cases

  • Monitor token consumption and costs in API calls.
  • Analyze usage efficiency across different models and providers.
  • Retrospectively review historical session data to optimize resource allocation.

Best For

  • Developers using the DeepSeek Harness Web profile.
  • Data analysts needing to control token costs.
  • Operations staff monitoring system performance and usage.