Plugins

Terminal-style AI-lab idle game where you hoard compute, train and upgrade models, publish papers, and prestige from neural nets to ASI.

Complete mobile PWA for DSH, built on dsh-mobile-gate: secure remote-access gateway plus install-to-homescreen (manifest + service worker), offline capability, touch gestures (pull-to-refresh, edge-swipe back, pinch-zoom font), agent-done push notifications, and touch-first layout — desktop unaffected.

Static and runtime security guard for dsh: rule-based scans for malicious code, prompt injection and token waste, runtime interception of dangerous tool calls, /scan command, plugin_scan tool, web panel, and allowlist.

One-command bundle of the three AgentFrame plugins: memory-director, compaction, and aura-scheduler together.

Update checker (npm edition): a settings page comparing local and latest versions of @deepseek-ai/dsh and every installed plugin, with a red-dot nav reminder and ready-made update commands.

Read-only health clinic for the installed DSH plugin set: loader health, dependency integrity, version compatibility, install-script risk, duplicates and patch integrity, delivered as a model tool, a Settings dashboard and JSON reports.

Boot-time guard that repairs corrupt session logs, preflights plugin composition, vm-checks client artifacts, smoke-tests host apply() and quarantines crash-causing bundles, so a broken plugin can never brick startup.

Task-completion notifications via ServerChan / DingTalk / Feishu / generic webhooks.

Drive your dsh agent from WeChat over the iLink gateway: two-way encrypted media transfer (auto-decrypt inbox + /send), proactive weixin_send tool, completion notifications, multi-session switching, and a health panel with file logging.

Conversation-completion notifications for DeepSeek Harness: Windows toast + sound on turn done / error / goal, plus ask & approval alerts - background-only, with a volume-adjustable bell in the conversation header.

Task-to-skill pairing meta-skill with a laziness ladder for minimal loading, a user-editable routing table, and a quarantine, SkillSpector scan, human-approval install flow.

Model-facing image_describe (识图) tool over the DashScope OpenAI-compatible API (qwen3.7-flash), plus a paste bridge: on text-only sessions, pasted images auto-convert to file paths at send time and render back in the transcript, so they never trip image admission. Bring your own DASHSCOPE_API_KEY; endpoint/model/budgets configurable, redirect-proof HTTP client, works in every agent preset.

Vision for any DSH route: paste images in the Web composer with intent-aware auto-analysis, delegate workspace image reads to a Kimi/MiniMax vision subagent, and materialize pasted originals for editing.

On-demand vision for text-only DeepSeek models: upload images, and the model calls a view_image tool backed by any OpenAI-compatible vision endpoint (Qwen/DashScope by default).

Local-first structured vision for text-only agents: images go to a local OpenAI-compatible VLM and come back as JSON evidence (summary, verbatim OCR, layout regions, entities/relations, colors, explicit uncertainty), with anti-hallucination fallback and an optional paste/upload bridge; zero cloud cost, images never leave the machine.

Synesthesia Encoder for DSH: a vision model translates images into compact structured spatial text (canvas/elements/percentage coordinates), giving text-only LLMs pixel-level image understanding via the `mm_vision` tool.

Routes chat images to a fixed OpenAI-compatible vision model, returns factual observations to the selected main model, and reuses session-scoped observations across replay, compaction, and restarts.

Measures design rules on the rendered page rather than in the CSS — type-scale count, WCAG AA contrast against the real composited backdrop, spacing-grid drift, palette size, tap targets, and the tells of generated UI such as violet gradients caught by hue angle; ships a design-system skill.

A `search` tool that learns to pick between fast and thorough search strategies using a contextual bandit (Thompson sampling), improving from real usage instead of a fixed heuristic.

Qichacha (企查查) enterprise data source as a dynamic Cordis plugin: 5 agent tools (auth with QR login, company search, knowledge search, tool search, tool execute) covering 100+ data dimensions (business registration, shareholders, judicial, IP), plus a settings-page account manager; network/file IO via DSH subprocess spawning node -e, credentials interop with the local BizOwl client.

HTTP/API debugging toolset: full-control HTTP requests (method/headers/body/auth), curl command parsing and one-step execution, in-session request history with side-by-side response comparison, pagination-aware Link headers, and DeepSeek-branded UI cards (optional dsh-http-tools-ui).

A slim port of Headroom: compresses tool output, cuts up to 20% of context, without affecting the model's context cache or agent performance.

Auto-discovers model listings, pricing and capabilities from OpenAI-compatible API hosts, normalizes them, and emits ready-to-use model configs for dsh.

Hana-style assistant manager for DSH: visual preset management (cards, yuan personalities, avatar cropping) plus a full Hana memory stack (rolling summaries, layered compilation, SQLite fact store, deep memory).