Plugins

A floating credit/quota widget for the DeepSeek Harness Web: shows DeepSeek / OpenCode Go balances and usage, and tracks DSH session token usage and estimated cost by day, month, and model.

Cross-session memory plugin for DSH (v0.6.1, npm @asherliner/dsh-memory-connect): auto-extraction, SQLite FTS5 indexing, RRF semantic recall injected into the system prompt every turn, scheduled maintenance, LLM-powered consolidation via ctx.llm, context explosion prevention, and global Soul identity via ~/.dsh/soul.md. v0.5.0 added temporal context graph (valid_from/valid_until/supersedes, append-only corrections via reviseMemory), trust model (recalled history injected as untrusted reference, current instruction always wins), turn-end summarization. v0.6.0 added local semantic embeddings (BAAI/bge-small-zh-v1.5 via embed_server.py, cosine + RRF fusion with FTS5, no cloud calls, ~232 msgs/sec on CPU). v0.6.1 fixed ~ expansion in path config. Zero-config install.

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

Learning Mode agent preset: a coding agent that teaches while coding — concrete scenario-grounded explanations, Socratic guidance, and TODO(你) practice blanks, modeled on Claude Code's Learning output style; installable via dsh plugin add (dsh-learning-mode on npm).

Compact Codex coding mode for DSH: light prompt guardrails against GPT over-explaining, with an editable prompt and native Codex-style tools.

dsh client bundle: a sidebar usage & cost dashboard — token stats, peak/off-peak aware per-model pricing, membership ladder, achievements and a daily heatmap.

Health checks for DSH plugins: manifest/patch/entry/build/pack/install verification, model-callable plugin_check, profile host-shadowing + BOM detection, environment diagnostics, and poison preflight.

Exports metadata-only DSH turn, model, tool, approval, retry, and error events to Tuning Engines for governed traces, policy evaluation, cost analysis, and Work Session review, with a disk-backed retry queue.

Model high-availability failover (quarantine, circuit breaking, probe recovery) and subagent orchestration (fanout/pipeline/supervisor) with bilingual settings UI.

Composer mic for the Web UI: tap-to-monitor live transcription and hold-to-talk, with host Edge TTS reply reading that streams while the model generates, echo-pause during reading, and tap-to-stop.

Knowledge-base-driven agent preset: wiki long-term memory, a web KB UI, a management page and optional UI polish.

NewAPI (OpenAI-compatible gateway) LLM provider: registers a `newapi` route with chat-only model discovery, auto-fills model parameters (context window, reasoning effort) from models.dev, and adds a Web settings section for the base URL and API key.

Use third-party subscription accounts in DeepSeek Harness: chat, image understanding, image generation, and image editing with the models your subscription covers, with available models and reasoning levels synced from your account; currently supports OpenAI/ChatGPT subscriptions, more providers planned.

LLM provider that runs the locally installed Claude Code CLI as the model backend, so requests go through an existing Claude subscription instead of a metered API key; native tool calls are bridged over MCP.

Wallpaper Engine local library, background layer, native Scene panel, and desktop-mode controls for the DSH desktop shell.

One-click enter/exit Plan mode for the DSH web input bar, a quick-click shortcut for /plan.

DSH 插件:图片与文件直达纯文本模型——图片保留原生附件体验,PDF/Office/压缩包/视频/音频显示为附件栏方块,点击发送时自动转为工作区路径,配合 dsh-vision-toolkit 粘贴即看图。A DSH plugin that delivers images AND files to text-only models as workspace paths: images keep the native attachment UI, other files show as square chips in the rail, paths append on send — pairs with dsh-vision-toolkit.

MCP server manager plugin for DeepSeek Harness: Settings → MCP page, OAuth (PKCE + dynamic client registration) or static-token auth, tools registered as mcp__<name>__*

MCP server installer and verifier for DeepSeek Harness: interactive init writes real patch entries into a profile, list and validate inspect them, and a CI-enforced verifier connects to every curated server.

Codex-style reasoning-effort slider embedded in the model selector.

OpenCode Go quota ring: click-to-cycle progress ring (5h/weekly/monthly) left of the model selector, colored by urgency, with reset countdowns on hover.

Document reading mode for DSH.

AX-first macOS computer use for DeepSeek Harness: accessibility snapshots with stable refs, background input, Unicode typing, menus, apps, set-of-mark screenshots, on-device OCR, verified actions, and optional Jev (TypeSafe System One) semantic guards — 1

Auto thinking-effort routing for dsh: a locally fine-tuned 7-question classifier (ModernBERT, 842MB, <1s CPU inference) picks low/high/max for no-intent messages; in-session retries escalate the tier (low to high to max) under exclusion constraints; the judge service fails safe to low if unreachable. A tier chip in the input bar shows the live tier and the last 20 judgments. All inference is local - task text never leaves the machine.