What's Changed

• docs: add official-repository declaration to READMEs by @PerryLink in #2
• docs: complete the plugin family table across five languages by @PerryLink in #3

New Contributors

• @PerryLink made their first contribution in #2

Full Changelog: v0.3.1...v0.4.0

[0.3.1] - 2026-08-27

Fixed

• Declare the web-client inject packages (@deepseek-ai/dsh-client-connection,
  @deepseek-ai/dsh-client-locale, @deepseek-ai/dsh-client-runtime,
  @deepseek-ai/dsh-client-ui-settings) as optional peerDependencies so the
  bundle composition is explicit and standalone installs stay clean.

[0.3.0] - 2026-08-26

Added

• Persist daily and monthly budget buckets across restarts.

[0.2.0] - 2026-08-23

Changed

• The Settings budget tab now honors config.refreshIntervalMs for its polling interval and config.warnRatio for its bar tone (both were previously hardcoded), and renders a per-day usage curve over config.historyDays, which is now actually consumed (it previously declared a limit no code read).

[0.1.2] - 2026-08-22

• Compatibility release for DeepSeek Harness 0.1.1-rc.2: all @deepseek-ai/dsh-* devDependencies moved to rc.2 (peers unchanged at >=0.1.0-rc.8 <0.2.0); dshWorkshop.compatibility.dshVersions set to 0.1.1-rc.2; the READMEs, AGENTS notes, and CI compat/typecheck pins synced to rc.2. No source changes required — the session/llm/typert/commands/client seams used here are unchanged in rc.2.

rc8 compatibility release

[0.1.0] - 2026-08-16

• Initial release: aggregated token/cost metering, budget caps with threshold alerts and over-limit policies, carbon footprint estimation, per-model latency benchmarks, and the /budget command.

Added

• Aggregated token/cost metering per model, session, and day from the session/event stream (assistant/message usage with request/header provider/model attribution), priced through the built-in USD-per-1M table merged with config.prices.
• Session/daily/monthly budget caps with warn-ratio threshold alerts (webhook POST + desktop-notification flag) and alert/block/degrade over-limit policies; block/degrade short-circuit the llm/stream waterfall with a corrective error finish (loop-built requests are frozen and cannot be rewritten, so degrade names the target model instead).
• Carbon footprint estimation via the token→carbon bridge (tokens × kWh/token × PUE × regional intensity).
• Per-model latency statistics with p50/p95 aggregation over a bounded window.
• budget Typert Remote namespace (budget/status, budget/setSettings, budget/unblock) + the browser Settings tab (usage bars, model breakdown, alerts, cap editors, unblock).
• /budget command (overview, models, unblock <scope>).
• budget/alert and budget/block session audit events (microtask-deferred past the session-append reentrancy guard).

Ported (upstream assets, Apache-2.0 — see THIRD_PARTY_NOTICES.md)

• LLM-Cost-Estimator-CN: price table (CNY per 1k, verbatim) and cost formulas → src/estimate/models.ts, src/estimate/cost.ts; operational USD table in src/estimate/prices.ts.
• Mode-Latency-Benchmark: benchmark vocabulary and percentile statistics → src/estimate/latency-stats.ts.
• AI-Carbon-Footprint-Calculator: GPU/region/PUE data and formulas + equivalences → src/estimate/carbon.ts.