Plugins

A `describe_image` vision tool for text-only models: images (local path, URL, attachment) go to a configurable OpenAI-compatible vision endpoint and only the returned text enters the session.

Vision bridge for text-only models: paste an image, get structured JSON evidence (OCR, layout, semantics).

Installs a dedicated AI novel-writing preset and workbench: revisioned local project assets, a compact side drawer, and native approval-gated single-file changes.

Free vision for text-only agents: built-in keyless vision chain plus pixel tools (Q&A, grounding, crop, pixel diff, colors, OCR, SVG trace, cutout, screenshots); paste an image to use it.

Chrome sidebar extension that lets DSH operate your browser directly, no vision capabilities required.

Vision for text-only models: paste an image and the model switches to a Vision Toolkit variant for image Q&A, multi-image comparison, long-screenshot OCR, screenshot-to-UI reproduction, element grounding, and pixel diff. No API key by default — images are processed by the author-hosted free service, 100 per machine per day; configurable to your own provider.

Chrome sidebar extension that lets DSH operate your browser directly, no vision capabilities required.

Lets a text-only agent watch a live screen or camera feed: every frame is scored by how much the picture changed, only the ones that matter are packed into one numbered sheet with an audio envelope, so the model looks at what actually changed instead of a screenshot every few seconds.

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.

DeepWatch's capabilities, installable into an existing DeepSeek Harness profile

Chrome sidebar extension that lets DSH operate your browser directly, no vision capabilities required.

Integrates TypeSafe Jev for skill and file selection, task supervision, shared-finding corrections, tool-output filtering, approval assistance, and recorded-stage navigation in DeepSeek Harness; all features are disabled by default.

Dispatch work to DSH agents from Claude Code or Codex: native subagent progress, in-host worker sessions with per-tier presets, and a multimodal bridge for vision and image generation.

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

Chinese government document processing toolkit: GB/T 9704 format check, auto-fix, content revision (red annotation + strikethrough), template generation, Markdown-to-docx, and document header/footer/page-number injection for 24 official document types.

Create editable Blender product scenes in DSH. Start from speaker, cup and lamp recipes or imported models; view saved results on the Lighttable, compare revisions, and adjust dimensions, materials, cameras and lights. Place static models on actual mesh surfaces in one saved operation, then save a complete static view while preserving existing cameras and animation. Repair repeating texture coordinates on planar custom mesh faces while preserving geometry. Preview both recesses of a timber half lap before saving. Arrays and mirrors remain editable. Render images or saved-shot videos and export an editable revision with its assets. English guides cover creation, modeling, photography, recovery and agent tools. Native geometry inspections, immutable revisions and an approval gate help preserve existing work. Supports DSH 0.2.1-rc.1 and the legacy host, with 20 tools for modeling, review and delivery.

Adds account login, API and custom Provider setup, model switching, image fallback for text-only models, and token/cost attribution to DeepSeek Harness 47f.

Third-party adapter that registers the upstream REA (Reverse Engineer Anything) MCP server into DeepSeek Harness as mcp__rea__* tools. 2.x provisions its own runtime (Node, the REA CLI, a JDK, Ghidra) under <DSH_HOME>/plugins/rea-bundle and launches REA through a shim, so a machine with nothing pre-installed works after one setup run; 1.x (branch v1, npm 1.0.6) is configuration-only and expects REA and Ghidra to be installed already. Analysis is upstream's: native binaries (Hopper/Ghidra/IDA), JavaScript/Electron apps, .NET assemblies, websites, Android APKs and adb devices, firmware, and process captures.

AI agent skills for Sealos — deploy any project, provision databases, object storage & more with one command. Works with Claude Code, Gemini CLI, Codex.

Cross-platform Computer Use for DeepSeek Harness: virtual-mouse human-like operation (screen_observe + computer_click + type + key + scroll + drag, 11 tools), AX-tree zero-vision-cost mode with free GLM-4V-Flash vision fallback, safety guards (snapshot TTL / app whitelist / dangerous-action approval / password-box protection).

A workbench for engineering and math problems, with nine tools for reading images, solving, plotting and verifying answers, plus four bundled skills. Needs Python with numpy and Pillow. Image reading, figure checking and formula typesetting rely on a vision provider you have configured a key for, so those images are sent to it, and without a key they fail while the other tools still work.

Daily game automation on Windows for Genshin Impact, Honkai Impact 3, Honkai: Star Rail and Zenless Zone Zero: a scheduled task per game drives BetterGI, the game-daily skill family, March7thAssistant or ZenlessZoneZero-OneDragon, with per-game switches, a foreground guard, in-run supervision that only steps in on trouble, and one report session per game that later results are appended to. Requires the DSH desktop app and the matching external engine installed.

Novel-writing guardrails: 24 novel_* tools enforce fact-ledger consistency, outline and chapter gates, a deterministic machine audit with a zero-token AI-flavor scan, and proposal-only revisions; a right-panel workbench covers reading, TTS playback, whole-book checks and batch drafting with checkpoint resume.

Design-fidelity QA for text-only models: a `deepseek_vision` tool borrows an eye from any OpenAI-compatible vision route, so the model can judge whether an implementation matches its mock — shipped with the benchmark behind that judgement (four fixtures, 23 injected defects, raw transcripts) and the questioning discipline it depends on.