dsh-usage-report

dsh-usage-report

Global token usage and cost dashboard for DeepSeek Harness, as a DSH bundle.

Every assistant message in a DSH session log carries its provider-reported usage (input / cache-hit / output tokens). This plugin reads all sessions in $DSH_HOME/sessions, aggregates them, and adds a Settings → Usage page with:

• cost / request / token cards for a chosen range,
• an hourly bar chart for a single day and daily charts for longer ranges,
• per-model panels with request and token charts,
• a custom date range picker,
• CSV export of the filtered range,
• a manual price entry form for models that are not in the built-in price list.

Install

dsh plugin --profile web add dsh-usage-report

Then add dsh-usage-report to dsh.profile.bundles in your profile's package.json (or install it from a plugin-market index, which does both).

Pricing

Costs are estimated, not billed. The built-in table follows DeepSeek's published prices, including peak/off-peak hours:

per 1M tokens	deepseek-flash off/peak	deepseek-v4-pro off/peak
input (cache hit)	$0.003 / $0.006	$0.022 / $0.044
input (cache miss)	$0.15 / $0.30	$0.66 / $1.32
output	$0.60 / $1.20	$1.98 / $3.96

Everything is overridable in $DSH_HOME/storages/usage/prices.json:

{
  "usdCny": 7.1,
  "tiers": { "flash": { "hitPeak": 0.006, "hitOff": 0.003, "missPeak": 0.3, "missOff": 0.15, "outPeak": 1.2, "outOff": 0.6 } },
  "models": { "my-model": { "missPeak": 0.5, "outPeak": 2 } },
  "providers": { "my-gateway": { "models": { "gw-model": { "missPeak": 0.5, "outPeak": 2 } } } }
}

Resolution order: providers[provider].models[model] → providers[provider].default → models[model] → built-in tier. A model that resolves to none is unpriced: its tokens are counted, its cost is excluded, and the Settings page asks you for a price.

Web search calls are billed separately by DeepSeek ($0.75 / 1000 successful requests) and are counted on their own.

Known limits

• Session logs record provider and model but not base_url and not which credential paid, so two routes of the same adapter cannot be told apart after the fact. Give each route its own provider (custom adapter) if you need them separate.
• Costs include only model usage and search requests; image tokens and any provider-side fees not represented in usage are not counted.

License

MIT