DSH Dashboard

Per-session statistics for DeepSeek Harness: token usage, timing, tool and skill activity.

Adds a Statistics tab to the conversation view. The numbers update live as the session progresses.

中文 | English

Compatibility

Built for DeepSeek Harness 0.1.2-alpha.2 (npm alpha dist-tag): the client half targets that release's conversation.view slot and tokenUsage / sessionStats projections, and the package's dsh.client manifest and peer dependencies declare the matching client packages. Only 0.1.2-alpha.2 is supported: earlier releases (e.g. 0.1.1-rc.2) have no useChat standard hook, so the tool and skill tables would silently stay empty.

Install

Releases are version-aligned with the harness: each version is built for, and named after, the matching @deepseek-ai/dsh release.

From npm

Pick the published version matching the harness you run:

dsh plugin --profile web add @wenqi_bian/dsh-dashboard

The npm package ships the prebuilt host and client bundles, so no build step is needed on install. The source's current version 0.1.2-alpha.2 is not published yet — install it from source below.

From source

The source version — 0.1.2-alpha.2, built for the harness source at the dsh-v0.1.2-alpha.2 tag — is not on npm:

dsh plugin --profile web add .

Run it from the repo root; it links the plugin into the web profile that $DSH_HOME points at. Rebuild after any code change with pnpm run bundle (see Quickstart).

Quickstart

pnpm run bundle                    # emit lib/index.js (host) + lib/client.js (browser)
dsh plugin --profile web add .
# restart dsh web, open a conversation, pick the Statistics view

The tab appears beside Chat / Trajectory and reads everything on the client from the session's tokenUsage / sessionStats projections and the conversation snapshot.

What it shows

• Token layer: input (fresh + cache read), output tokens, cache hit rate.
• Timing: LLM time, tool time, first-token latency, decode rate.
• Tool activity: per-tool calls, fails, total and average duration, drawn as proportional segments with a four-color cycle.
• Skills: which skill tools were invoked, as chips.
• Insights: tool success rate, think/tool time split, tokens per turn, input:output ratio.
• A zh / en toggle in the top-right.

Data source

Stats are derived on the client from the conversation snapshot's tool-result nodes and the built-in sessionStats / tokenUsage projections, the same source the shipped composer StatsLine uses. There is no host @Remote call, because @Remote is not generated for a runtime-loaded out-of-repo plugin. The workspace cross-session aggregate is not implemented.

License

MIT.