dsh-usage-cost
Run the following command in DeepSeek Harness:
dsh plugin install fflow2023/dsh-usage-cost
Paste the following prompt into your AI chat to install this plugin:
Run dsh plugin install fflow2023/dsh-usage-cost in your terminal to install permanently; the full source is available at https://github.com/fflow2023/dsh-usage-cost and the plugin takes effect after restarting DeepSeek Harness.
About this plugin
When you hammer DeepSeek models all day through DeepSeek Harness, the token counters at the bottom scroll by fast, but nobody tells you what it actually cost. dsh-usage-cost fills that gap: it appends a running cost figure with a peak/off-peak split right next to the token stats in the status bar, and adds a global dashboard in Settings that breaks down every historical session by model and time-of-day.
The implementation is deliberately minimal: a single session projection unit that folds over the existing event log in O(1) per event, so state survives restarts and cold reads. All price constants live in one object; when DeepSeek updates rates you edit that one file and bump the version. Pure JS, zero build steps, zero runtime dependencies — clone and go. It injects no prompts, triggers no extra requests, and has zero impact on KV Cache or token accounting.
Who is it for? Developers and teams running deepseek-v4-pro or deepseek-v4-flash for daily development or batch inference who want a live per-session cost readout and a monthly big-picture ledger. It is read-only and does not alter any model behaviour, so it slots into your existing workflow without friction.
Use Cases
- Glance at live session cost and peak/off-peak split in the status bar
- Review global cumulative spend broken down by model in Settings
- Compare peak vs off-peak unit prices to spot expensive call windows
Best For
- Individual developers building daily with deepseek-v4-pro or v4-flash
- Team engineers keeping a monthly API budget in check
- Harness users who want a live cost readout without changing their workflow
Related Plugins
ModLens is a vision plugin for DeepSeek Harness that gives text-only models sight by reading images pasted directly into chat, with zero-config setup and multiple vision engines.
On-demand vision for text-only DeepSeek Harness agents: built-in free keyless vision chain and 14 vision tools, routing image turns as tool calls to vision models with pixel fidelity, no Python needed, one-command install.
Give text-only models in DeepSeek Harness eyes, enabling image Q&A, long-screenshot OCR, UI restoration, and GUI visual tasks.
An armor-breaking plugin for DeepSeek: rewrites instruction priority, expands refusal-blocked phrases, persists across sessions, and shows a green active indicator.