Plugins

Automatic Flash/Pro routing for DeepSeek Harness subagent calls: light tasks run on a cheap model (e.g. V4 Flash), heavy tasks stay on the strong model — with per-call overrides and /delegate session modes.

Econometrics research assistant: 6 tools covering method selection, data preparation, model specification, empirical analysis with Python/R/Stata code templates, robustness checks, and result reporting.

Switch the current DSH session's model with a natural-language sentence, on web, TUI, or IM bridges (WeChat, Feishu), without touching the panel.

Multi-agent collaboration suite: user-configured specialist roster, persistent on-demand dispatch (team_call/team_message/team_status/team_close), clone instances, star-topology relay, model comparison and a multimodal vision bridge.

Live llmtrim savings dashboard in the DSH web UI: Settings page with paid, would-have-cost, saved-today and saved-this-week cards plus per-model breakdown, and a configurable rotating or static stats strip under the composer.

Shows the selected model provider's account balance next to the composer's model seat once a model and reasoning effort are selected.

Cross-platform Web UI launcher for DeepSeek Harness: webui_start/stop/status/open model tools, a /webui start|stop|status|open slash command, a Settings-page launch card, and an optional desktop shortcut that starts the Web UI hidden and opens the browser once it is fully ready (dsh default icon).

SSQ (China Welfare Lottery Double Color Ball) helper for DeepSeek Harness: 1000-draw trend tracking, six prediction models with ensemble voting, structure analysis and dan-tuo/random generation via the ssq tool.

Adds a "用量" section to the DSH Settings page that shows your MiniMax Token Plan usage (5h and weekly windows with reset countdowns), per-model breakdown, and video bonus credits; fetches `/v1/token_plan/remains` via a trusted plugin that holds the API key, auto-hides when no key is configured. MIT.

World-model research toolkit: inspect frames, name 3D / pixel / latent routes, score pred vs GT, and RSI skills / wm.yaml.

Model-facing `generate_image` tool for text-only models: the model asks for a picture in natural language, Gemini draws it via the Antigravity CLI, the image is saved to a configurable output directory, and the tool returns the file paths for the model to use in its work.

DeepSeek Harness plugin: generate_ui — the model turns a request (+ data) into an A2UI document rendered as an interactive surface in the chat, generated through dsh's own model service (valuz-genui core)

OpenCode Zen free models provider for dsh

Migrate Codex CLI conversations, MCP servers, memories and project files into DeepSeek Harness sessions, workspaces and MCP registrations. 把 Codex CLI 的对话、MCP、记忆与项目文件迁移为 DSH 会话、工作区与 MCP 注册。

Rust dev domain plugin for DeepSeek Harness (dsh): verification pyramid fmt→clippy→build+test→miri+audit(+semver) as model-facing tools + graph-layer governance (anti-Goodhart 8 rules incl. #[allow] density / deleted tests, blind-spot via Cargo.toml & pub

Linux kernel dev domain plugin for DeepSeek Harness (dsh): verification pyramid L1-L4 as model-facing tools + graph-layer governance (anti-Goodhart / blind-spot / patch-conflict) + AGENTS.md contract as a system-prompt section. Plain JS, no build step.

DeepSeek Harness plugin: import models with capabilities from an OpenAI-compatible /v1/models listing (gateway capability extensions) and edit per-model capabilities (thinking, vision) on its settings page

Pokemon-style catch-em-all: your real work is the tall grass — turns, tool calls and errors spawn wild encounters; throw balls, fill a 28-slot dex, walk a team of six. Zero tokens, invisible to the model.

Pull a 3D lever next to the send button and turn the model's reasoning signals into a pixel-art gacha reveal.

Hands standalone jobs to the Cursor CLI signed in on the same machine, with CLI readiness and proxy diagnostics, child-model settings, and a read-only connectivity test.

Plugin compiler with a blueprint registry (scaffold → validate → deploy) exposed as a runtime service, three model tools, and a blueprint-gallery settings page.

Reads assistant replies aloud via Fish Audio API only (bring your own key): per-message read-aloud, auto-read toggle, and a settings page for model, voice reference_id, encrypted API key, and proxy.

Chat image-attachment bridge with a `view_image` tool for any OpenAI-compatible VLM (local Ollama or cloud): pasted/dropped images become `view_image` path markers before reaching text-only DeepSeek models.