Plugins

Exact-profile control of your real logged-in Chrome browser: an MCP server plus MV3 extension and authenticated localhost bridge for tabs, locators, file injection, media export, and durable scroll capture, with no debug profile or remote-debugging port.

Ingestion-time tool-output curation: keeps error and warning lines plus head and tail, spills the full text to disk with a pointer the model can read back; covers every text tool, small results pass through untouched.

Gives an agent its own email inbox through AgentMail — eleven tools for sending, reading and labelling mail, plus inbound mail delivered into one harness session per email thread, with an approval gate and recipient allowlist on everything outbound.

Apportez votre historique Claude / ChatGPT existant dans DSH : import instantané sans tokens du memories.json d'une exportation Claude, transcriptions locales Claude Code sans export nécessaire, et conversations distillées via le modèle déjà configuré dans DSH. Chaque mémoire porte des preuves verbatim vérifiées contre la source par du code — une paraphrase est écartée et comptée, et une preuve trouvée uniquement dans les tours de l'assistant est toujours orientée vers la revue plutôt que stockée.

Model-authored context pruning for DeepSeek Harness through the official compaction API.

xAI Grok subscription chat: SuperGrok or X Premium+ OAuth, a displayed model subset, Responses through the Grok CLI proxy, and always-on server-side web_search and x_search.

Prise en charge d'un fournisseur Claude personnalisé pour DeepSeek Harness.

Plugin fournisseur DSH pour l'accès à Codex via un abonnement ChatGPT.

Plugin DeepSeek Harness : outils de navigation web (markdown / capture d'écran / pdf) propulsés par Cloudflare Browser Run — véritable Chrome headless avec rendu JS, sessions de connexion et prise en charge WebMCP.

Saisie vocale via la Web Speech API du navigateur : zéro serveur, zéro clé, zéro téléchargement de modèle (Edge=Azure, Chrome=Google speech).

Barrière de point de douleur imposée : après deux expériences non convergentes, elle injecte les trois questions, refuse les appels d'outils non exploratoires tant qu'elles ne sont pas répondues, et bloque les nouvelles tentatives dans la même direction.

浏览器控制插件：把 browser-use / browser-harness 的 Browser Harness 注册为 DSH 会话工具，让 agent 直接写 Python 驱动真实 Chrome（导航/点击/输入/取无障碍树/截图），支持独立专用浏览器与附着系统浏览器两种模式；配置 TypeSafe key 后条件启用 browser_autopilot（Jev 目标式自动化）。

Adds a speaker button immediately right of the Like button on every finalized assistant reply; click it to hear the reply through the browser speech engine, and hover it for a speed and voice panel.

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.

Mounts the mcptoon MCP gateway as one stdio server, so the agent discovers the tools of every MCP server you have configured through mcptoon's compressed view instead of full JSON schemas on each turn. Ships a doctor CLI that checks the mcptoon executable resolves and prints the exact command/args override when it does not.

Redraws line-level diffs on edit, write, insert and str_replace_editor tool cards — unchanged lines render once and changed characters are underlined — covers run_code (PTC) sub-calls, adds a per-turn changed-files card, and opens or copies any of those paths from the card rows and the tool rows through a host route the plugin serves itself.

Adds a Run button to shell code blocks in replies, executing the snippet through the host shell service and showing its output in a floating bottom-right dock, or in one bottom-panel tab per run.

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).

Manual model probe for DeepSeek Harness: a session-view tab where you pick provider, API type, model and route, then send one real raw HTTP request over openai-completions, openai-responses or anthropic-messages — streaming or not — and read latency, TTFT, HTTP status, token usage and a three-layer diagnosis chain (API request / model response / strict check) on one screen, with errors in plain language.

Model-autonomous context compaction for DeepSeek Harness: a mind_map tool surveys the agent context surface (stable seq ids, roles, token weights, clear-boundary markers, the clear-mind playbook), a clear_mind tool replaces a chosen conversation-history span with a model-written checkpoint through the native compaction transaction, call and result pairs self-collapse into one-line tombstones, and a proactive reminder nudges the model when the context grows long or a turn runs too many steps.

Quick Control panel for DSH Web: a docked or floating shortcut panel whose switches other plugins register through the quickControl service. Includes a theme and skin switcher plus fine-grained NO_PROXY proxy controls, and works with dock-base or standalone.

A draggable whale-maid physics widget for the DSH Web UI: throw it around, catch token coins, and eat rice bowls.

Mobile-first Web UI quality-of-life pack for the DeepSeek Harness: a Chrome-style active-session tab bar, full-screen sidebar swipe open/close, overlay sidebar, collapsed-sidebar recents, keyboard-suppressing session switching, viewport/IME keyboard adaptation, touch feedback, code/table inner scrolling, hidden permission dropdown, a full-screen settings rewrite with tab memory, and a compositor-driven status pulse, no-touch-drag long-press protection — 14 features, each independently toggleable from the Settings QoL section with instant effect, persisted per browser in localStorage; pure client-side CSS/JS, mobile rules locked to viewports of 768px and below so desktop stays untouched.

A full-page console panel opened from the sidebar, with eight tabs: an overview; a session list read from the sessions service (title, directory, running state, updated time, with filters); installed bundles and Loader entries from pluginInventory and pluginManager, with per-bundle enable switches; a plugin market listing public topic:dsh-plugin repositories from the GitHub search API; a model and provider editor that writes the llm-pi-ai settings namespace through remote.settings.mutate (switch a session model, add or edit providers and their model lists, and store API keys via remote.credentials, merging over the stored profile so fields the form does not expose survive an edit); MCP and skills tabs; and a usage tab that reads the host's loopback /api/dsh-usage/overview document. Services are declared through ctx.inject, so a deployment that omits one mounts without the console rather than crashing.