Plugins

Render and iterate on Blender scenes from DSH: 16 tools for scene specs, previews, visual review and delivery renders, with immutable revisions and an approval gate.

Same-origin in mechanism with the two loudest lines in context compression. The code-skeleton gate follows the skeletonization approach of Headroom (Apache-2.0), whose published headline is 20% fewer tokens for coding agents and 60–95% fewer tokens for JSON, same answers. The estimator channel follows TokenPilot (arXiv:2606.17016), which reports up to 60% lower cost for long-session agents. Both figures are theirs, quoted as-is; this plugin ships no benchmark of its own and claims no reduction of its own. What it adds for DeepSeek Harness: choose a compression profile, set the Auto Compact trigger level and toggle code-skeleton compression from one settings section, with exact DeepSeek V4 tokenizer measurement, same-revision count verification, and fail-open behaviour that keeps the original tool results on unsupported models.

An editable diagram canvas in the DSH sidebar plus two agent tools (diagram_read, diagram_apply) that read and edit the workspace in place, using native .drawio files: the canvas opens and saves them losslessly — cells it does not understand are preserved byte-for-byte — and a file-fingerprint revision powers optimistic locking against concurrent edits.

DeepSeek Harness plugin: auto-route image-bearing requests to deepseek-v4-flash-vision-exp, then fall back to the original model.

Desktop computer-use for DSH: captures the Windows screen into the agent vision channel and drives mouse/keyboard through a Codex-style overlay with Esc abort; auto-detects Windows native or WSL hosts.

Windows computer use forked from computer-user: 13 computer_* tools that read the screen and drive the mouse and keyboard. Image-capable routes get the screenshot as a real image with an exact image-to-screen mapping, so no external OCR; elements return as UI Automation refs so a click lands on the exact control rectangle; Ctrl+Alt+Esc stops every call.

Model picker for the DSH web composer that replaces the model seat: a search-first popup whose left column indexes the suppliers and scrolls the grouped list, with a favorites group pinned at the top, image icons separating native from bridged vision input, and a reasoning-effort panel opened from the right zone of the trigger.

Evidence-first attention supervision for DeepSeek Harness with C0–C3 policy, deduplicated Inbox alerts, quiet hours, and redacted outcome records.

Provisions the kimi-webbridge bridge daemon (auto-install with SHA-256 verification, auto-start) and registers a 25-action skill that drives the user's real browser through the local bridge.

OpenAI-completions-compatible adapter for custom gateways (vLLM / LM Studio / self-hosted proxies): always role:"system", thinking driven by the model config, Qwen-style response split, and vision-model image input (single / multiple).

Transparent multimodal routing for text-only models: every image in every model call is fully transcribed (verbatim OCR, data, uncertainty zones, injection-hardened) by your own multimodal understander, with focused re-look via vision_relook and automatic retry on image-related failures. No bundled endpoints, no borrowed logins.

Free vision OCR with adaptive tile recognition for long documents and Markdown/Word/PNG/Excel export.

Per-profile settings overlay for DeepSeek Harness: global settings.yaml stays the baseline while each profile overrides any namespace through its own profiles/<name>/settings.patch.yml — object sections merge recursively, scalars and arrays replace wholesale, and !unset masks an inherited value. The overlay is transparent to existing plugins (they keep reading ctx.settings unchanged), writes land in the profile overlay only, and official schema validation, revision semantics, expectedRevision conflicts, watchers and events are untouched. Ships a settings command family (get/set/unset/mask/unmask/promote/demote/migrate/diff/layers) plus a Profile Settings section in the web Settings panel over a loopback RPC channel.

DeepSeek Harness vision plugin: 8 analysis modes (describe, OCR, chart data, UI review, object detection, compare, code-gen, debug), any OpenAI- or Anthropic-compatible vision API, with a built-in free vision model and automatic rate-limit failover.

Screen capture and external vision recognition: take_screenshot, list_windows, analyze_image and view_image tools with a configurable GPT vision channel (gpt-5.5 / gpt-5.6-sol / gpt-5.6-terra), API key via the credentials service, and a settings card; view_image shows the screenshot in the Web UI while the model context keeps text only.

Inline-Bildvorschauen im Chat für eingefügte Bilder, die über dsh-tool-vision an textbasierte Modelle überbrückt werden (Klick zum Vergrößern, alte Nachrichten erhalten ebenfalls Vorschauen).

Stufenweises Modell-Routing für DeepSeek Harness: think/build-Modellstufen, Vision-Delegation (das Vision-Modell beschreibt nur Bilder; das Arbeitsmodell handelt), Bildgenerierung und ein High-Impact-Guard.

Überwachung langlebiger Dienste (Dev-Server, Watcher, Mock-APIs) durch einen losgelösten Broker, die Turns, Sitzungen und DSH-Neustarts überstehen – mit einem Service-Tab in der Seitenleiste (registriert über die Erweiterungs-API von better-sidebar) und den Modell-Tools service_start/stop/restart/list/logs.

Design Studio-Tab: Design-Briefings werden zu html/css/js-Mockups mit Live-Vorschau, Element-Picker, Design-Agent-Chat mit Vision-Review, Identitäts-Presets und Zip-Export.

DeepSeek Harness (DSH) Web-Plugin: Dateien in den Composer einfügen, Ablage im Posteingang, Miniaturansichten. Text über read_inbox; Bildverständnis erfordert einen eigenen Vision-Skill (mitgelieferter image-reader + ARK_API_KEY).

DSH-Plugin für Dual-Modell-Routing: Die Hauptkonversation bleibt bei DeepSeek, Bildaufgaben werden automatisch an das explizit konfigurierte Vision-Modell verteilt und das Analyseergebnis geht zur Fortsetzung zurück an DeepSeek.

DeepSeek Harness-Plugin: eine `deepseek-vision`-Route, die Bildeingabe deklariert und eingefügte Bilder über die kostenlosen Zhipu GLM-Visionsmodelle transkribiert, bevor sie an den DeepSeek-Adapter delegiert.

DeepSeek Harness Out-of-Tree-Plugin: gibt einem Text-only-Coding-Modell Augen, indem es Bilder über ctx.llm an eine Qwen-VL-(DashScope-)Route leitet und Text zurückgibt.

DeepSeek Harness settings-pro Plugin: IM-Bridge (Telegram/WeChat), DeepSeek-Nutzung, Cross-Restart-Memory, Desktop-Pets und eine Vision-Bridge