Plugins

Model settings panel for DeepSeek Harness: set the default model for new sessions, plus separate provider/model/reasoning-effort overrides for subagents and goal rounds, with live reload.

Brings WorkBuddy free/paid models (domestic + international) into DSH without the desktop app, published as @bushizhanghan/dsh-workmate. This package is a maintained fork of Faide-cyber/dsh-workbuddy (whose upstream entry is already listed in this catalogue); do NOT co-install both, as they register the same provider id "workbuddy-ai" and will collide. v1.2.7 relocates over-long system prompts past the domestic gateway channel-risk classifier, rewrites the DSH brand string, and adds a debug dump; supports dsh 0.2.0-rc.2 through 0.2.1 (incl. prereleases).

Usage panel on the blank-session hero: sessions, messages, total tokens, active days, peak hour, favourite model and a contribution heatmap, over all time, 30 days or 7 days.

Replaces the composer's model and reasoning-effort selector with a two-level menu and six slider skins.

Next-prompt suggestions: after each answer it predicts your next message as grey ghost text in the composer. Tab fills it without sending, and the prediction model is selectable.

Generative UI: the model calls render_ui with a React snippet and the answer renders as an interactive card in the conversation. The snippet runs in a sandboxed iframe with an opaque origin, so it cannot read the host page, its storage or its cookies, and the frame CSP blocks the network; a message from a card is only sent after the user confirms it, and a failing card shows its error with the source behind a toggle.

Long-term memory for DSH stored as plain Markdown. Real user/assistant exchanges are captured into daily session files without a model call; an LLM pass then distils them into memories that are injected into the system prompt each turn. Retrieval is local (CJK bigram + BM25) in three levels — index, summary, full text — so the model only pays for the detail it asks for. Memories carry priorities; P0 and user-written entries are never changed automatically, and forgetting marks an entry stale instead of deleting it, so the whole chain stays auditable in status.md. Six tools cover recall, write, status, extract, feedback and the skill library; when a cluster of memories is ready to become a skill, a card appears at the end of the turn, writes the SKILL.md on click, and installs it only after you approve. Learning is on by default and follows the conversation model; turning the master switch off stops the LLM entirely while capture keeps working. Failures are surfaced in the prompt and in status.md instead of being swallowed. Pure JavaScript, no third-party runtime dependencies beyond the harness packages.

Diagnoses third-party LLM API endpoints from the sidebar: a four-layer check (DNS/TCP/TLS, key and model, a real streaming conversation, then timed repeat runs) reporting time-to-first-token, output speed and success rate, with every failure attributed to the step it stalled at and the byte count that actually left the machine. 37 provider presets, no model names hardcoded.

Runs one-time and recurring Agent tasks in fresh DSH sessions, with natural-language schedules, visual schedule editing and session-linked history. Supports custom models, skills, pause on failure, and permission settings.

Lighten the load on your agent: one-click toggles for skills and MCP, so the context holds only what this turn actually needs. Skills and MCP servers you switch off are no longer injected into the model context — cleaner, and lighter on tokens.

Cross-session memory kept as a short Markdown living document that is rewritten in full at the end of every turn, written to by model tools, the page, and automatic rewriting alike.

Set an earlier automatic-compaction threshold per model on DeepSeek Harness: /autocompact writes a session-only override keyed by exact provider/model (off falls back to the backend defaults, default clears it), a Settings → Autocompact page manages the persisted per-model thresholds every session inherits, and a composer status entry shows the current reading.

Expose every free model from a local opencode serve catalog — all providers, including your own custom ones — to DeepSeek Harness's model picker, with OpenCode's native tools denied so execution stays in DSH.

Unofficial Volcengine Ark Coding Plan LLM provider: registers an `ark-coding` route with 12 models (capacity facts measured per model), multi-account with automatic rotation, a bundled pinned arkcli for the official model catalog and live plan quota, plus a settings wizard and a sidebar usage card.

Per-turn session cost chip in the composer stats row: each turn's spend with the model it used, the conversation total split into cache-hit and cache-miss (CNY), priced from an official price table that refreshes on demand.

Plays local video and audio inside the conversation. Media files the model writes during a turn - screen recordings, transcodes, voiceovers - get a player under its final reply automatically when their paths appear in its commands, the command output or the reply; the play_video tool plays any other file. Video shows as compact cards with a hover preview, audio as a slim seek bar, streamed over HTTP Range so long files seek instead of loading into memory. Also registers a sidebar media preview renderer, since the built-in previewer treats mp4 as unviewable. The look_video tool lets the model see a video itself: a hidden local Chrome or Edge grabs frames into a time-labelled contact sheet (or big frames at given times, or a short range frame by frame), checks the audio (silence, too quiet, clipping, one channel only) and the picture (black or frozen frames), and reads MP4 / MOV headers for phone compatibility (codec, HDR, faststart and more), with ffmpeg fixes.

Agent plugin for major harnesses — DeepSeek Harness (DSH), OpenCode, Pi and Codex: define custom AI agent roles with per-role prompts, models, skills, permissions, memory and multi-agent orchestration.

Local Git plus the four GitHub remote tabs in the official DSH right sidebar: Changes (per-file staging, line diffs, and commit messages written by the model already configured in DSH), History, then Code, Issues, Pull requests and Actions proxied by the host half. The token stays in the host half through ctx.credentials and /dsh-git/* accepts loopback requests only.

An OpenCode Go model source for DSH: it reads the opencode-go catalog from Models.dev and serves it on a separate opencode-go-live route, leaving the built-in opencode-go route in place. The catalog snapshot is cached under the DSH home and a failed refresh keeps the last successful one; its own Models settings card edits the route key and can render OpenCode Go subscription usage.

Model failover and turn keep-alive for DeepSeek Harness: a request that fails with a model error before any visible output falls through the models of the configured provider groups (then the next provider) until one streams; transient failures retry with backoff under a liveness watchdog, a dead turn self-resumes with templated 「继续」 wake-ups, and an optional full-auto mode auto-answers permission, question and plan approvals with every decision appended to an AUTO-MODE.md audit log.

Generate images from the agent on the user's own GPU by queueing a workflow on a local ComfyUI install over its HTTP API: Z-Image-Turbo for fast drafts and Qwen-Image-2.1 for final assets and legible in-image text. Registers comfyui_generate and comfyui_status plus a skill covering model selection and prompt craft, has no runtime dependencies, and starts ComfyUI headless when no server is listening.

Recall a message you already sent: the session log stays append-only and nothing is deleted, while the recalled turns leave the model context and the transcript, the text returns to the composer, and the files those turns changed wait in a quarantine you can review and clear from its own settings page.

One authenticated OpenAI-compatible endpoint (built on CLIProxyAPI) for dsh and other tools: gathers WorkBuddy, OpenCode Zen, local Ollama / LM Studio and CLI subscriptions such as Codex and Kimi, signs in from the dsh page without installing their CLIs, and syncs the picked models into dsh grouped by source. The settings page has a switch per source with free / paid tags, plan usage, API keys with 24-hour stats, per-source model picking, and update / uninstall. Runs on macOS, Linux (including next to dsh-vps) and Windows.

Stops DeepSeek Harness from compacting (summarizing) your conversation far too early. Without it, compaction fires around half-full on smaller context windows; with it, compaction waits until the window is as full as you want (80% by default), settable per provider/model. It only changes when compaction happens — what a summary keeps is still DSH’s decision.