Plugins

Echtzeit-session-Token-Nutzung und geschätzte Kosten in der composer-stats-Zeile, wobei das pricing von der offiziellen pricing-Seite von DeepSeek abgerufen wird.

Online-Statusanzeige ohne Tokenverbrauch für die DeepSeek Harness-Web-UI: ein immer sichtbarer Online/Offline-Punkt in der Konversationskopfzeile, automatisch alle 15 Sekunden aktualisiert ohne LLM-Aufrufe.

DeepSeek-Harness-(DSH)-Plugin: ein „Denksprache“-Selektor unter Einstellungen → Allgemein, der dem Modell über einen System-Prompt-Abschnitt mitteilt, in welcher Sprache es denken soll (Chain of Thought). Keine zusätzlichen Aufrufe, keine Latenz, 12 Sprachen.

Command-Code-Go-Provider-Adapter für dsh: durchsucht den Go-Plan-Modellkatalog dynamisch und bedient /alpha/generate über die LLM-Schnittstelle des Harness

Langfuse-Observability für DeepSeek Harness: exportiert Sitzungsprotokolle als OpenTelemetry-Traces (GenAI-Semantikkonventionen) an den OTLP-Endpunkt von Langfuse

Automatisierte Unit-Test-Generierung: ein `/testgen`-Befehl und ein generate_tests-Tool, die Tests erstellen, ausführen und reparieren, bis sie bestehen (LLM plus Offline-Template-Generatoren; vitest/jest/mocha/node:test).

Adaptives Modell-Routing: Komplexitätsklassifizierung pro Anfrage mit automatischem Anbieter-Routing.

只填 Key 的服务商预设：50 家渠道为 DSH 一键写入官方 llm-pi-ai 路由与凭据（插件），并可写入腾讯 WorkBuddy、OpenCode（Key 走其 auth.json）与 cc-switch 的配置（CLI）；核心为纯 JS，可被更多应用的外壳复用。

Ship DeepSeek Harness to your own VPS: one command (or a form in DSH Settings → VPS Deploy) installs a zero-dependency login gateway, automatic HTTPS through Caddy, systemd supervision and one-click DSH upgrades.

Keeps a project's documentation in step with its decisions on DeepSeek Harness: the same append-only decision log and derived design snapshot, kept fresh in-process instead of by a hook process.

DeepSeek Harness 代理插件：在设置中配置全局 HTTP(S) 代理，或为单个 LLM 提供商单独配置代理 / 直连

Use ChatGPT Codex, Claude, Grok, Antigravity, Kimi, GLM, Cursor, Kiro, Copilot, Qwen, ERNIE, Spark, JetBrains, Perplexity, Replit, and Cody accounts as DeepSeek Harness LLM providers via OAuth.

Windowed, reversible context management for DeepSeek Harness; replaces Basic compaction.

One-command VPS deployment for DeepSeek Harness: a zero-dependency login gateway with automatic HTTPS that proxies the native web UI, a token-protected setup wizard, and one-click DSH upgrades with automatic rollback; as a plugin it adds Settings → VPS Deploy with version, upgrade and gateway status, or the one-command setup when DSH is not yet on a VPS.

Markdown visual enhancement and session workbench for DSH Web Composer: live syntax presentation with caret-row source reveal, deterministic local completions and diagnostics; long pastes fold into persistent chips with inline context or file-mode delivery (@path across 50+ extensions) plus sidebar LLM rewrites (stop, preview diff, undo/redo); real-time context map with token gauge, composition breakdown, per-turn growth chart with anchor lines, in-place /compact, and turn forking.

OpenAI-compatible vLLM client (default :8000).

Session-aware local notes for DeepSeek Harness: capture notes and page selections in a floating Apple Notes style panel, let the LLM classify topics, auto-inject convention notes into the system prompt, and dispatch todos with recall context to active sessions.

Replaces the built-in llm-pi-ai and llm-deepseek adapters, the model selector and the Models settings page with self-maintained versions, tracks upstream @earendil-works/pi-ai through a safe-zone download with a one-click enable toggle, and adds provider quota and balance lookups plus a provider management page.

Multi-provider LLM API balances under the chat input, with presets for DeepSeek, OpenRouter and SiliconFlow.

Security-audit facts AI coding agents cannot compute, plus a verdict gate: euthyna history attributes deleted lines to commits and flags those from security fixes (--origins finds the first introducer), euthyna coverage reports which changed symbols no test ever invoked, and euthyna gate checks each finding against six gates, downgrading any without evidence to an observation.

Adds an Explain button to the approval card: one model call returns exactly three lines reading the pending operation, covering what it does, which files or system state it reads, writes, modifies or executes, and a risky/attention/safe verdict with a one-sentence justification. It also renders the detail area for `write` and `edit` calls, which the shipped renderer leaves blank because it only reads a `command` field.

TypeSafe Jev (System-One-Entscheidungsmodell) für DeepSeek Harness, durchgängig: jev_decision (Choice/Score/Noul, parallel, typisierte Antworten + Konfidenz), eine Karte Einstellungen > Plugins für API-Key und Schalter, Inline-Tool-Ansichten, die jede Entscheidung in der Konversation zeigen (Antwort, Konfidenz, Latenz, Kosten), ein Auto-Guard (deterministische + Jev-Risiko-/Loop-Prüfungen), ein lokales /jev-Dashboard und ein Online-Verifikationsbenchmark (96,3 % bei 27 gelabelten Fragen, Median ~283 ms auf jev-latest, 2026-09-21). Ehrlich ausgelegt: nur echte API, kein Mock-Fallback.

Fronts stdio MCP servers with three router tools (list_capabilities, describe_tool, call_tool) so schemas load on demand instead of riding every request; shapes every tool result through a deterministic output ladder that archives the original first (retrieve_spill reads it back); prices tool groups in tokens and can hide built-in groups per agent. Bridges to a standard-library Python router, so Python 3.9+ is required.

Compaction fallback for compaction-basic: when its LLM summary throws, comes back empty or cannot fit the window, the span is compiled deterministically with a seq pointer on every line (vendored dsh-compaction-instant compiler). Adds recall and search tools, /recall and /context commands, an idle trigger at 45% of the window and an on-disk span archive. The engine mounts as one row in the agent preset (docs/dsh-integration.md).