Плагины

Auto-switch the DeepSeek route to the vision model on demand: flash main session switches (A), pro keeps deep reasoning and delegates image reading to a vision subagent (B), subagents always switch, with fatal-failure fallback. No manual model switching.

Force the agent's chain-of-thought, planning, tool-invocation reasoning, and internal deliberation to stay in Chinese — the language constraint is injected at the top of the system prompt before every reasoning step, so it cannot be diluted by tool-returned content. A settings panel lets you edit the constraint text on the fly.

Tiered compaction hub: replaces the single 80% compaction cliff with a configurable ladder (default 30/50/70/90) whose tiers escalate from gentle to maximum consolidation laws, folds a deterministic keyword index into every checkpoint so facts accumulate across cycles, and summarizes through a model chain — primary plus up to 3 fallbacks, each with its own reasoning effort. Adds a Compaction module sub-tab and a Settings card; toggling off always restores stock compaction.

Replaces the official Models settings page with provider lifecycle management, per-model reasoning effort tiers, retry overrides, drag reorder and upstream model sync.

DSH plugin: auto-detect and configure model capabilities (reasoningEfforts + input modalities) for llm-pi-ai. Successor to dsh-reasoning-efforts.

Searchable model selector for DeepSeek Harness (DSH) web: a faithful fork of the native model seat (Model/Effort, provider-grouped list, reasoning effort) with a search box at the top of the model list that filters while you type.

Live reasoning-token badge — Think(5.2K) — on every Think disclosure row of the DSH web UI, ticking while streaming · 在 web 界面 Think 行标题后实时显示思考内容 token 数的 DeepSeek Harness 插件

dsh-think-autoexpand：让 DeepSeek Harness 里的 Think 思考行始终可见，并在流式输出时自动展开，实时查看 Agent 完整推理过程；保留工具卡折叠体验。

DSH plugin: per-model reasoning levels for custom API models, with family presets and an official-settings-page editor.

Per-model reasoning effort (off/minimal/low/medium/high/xhigh) for GPT, Codex, Gemini and OpenAI-compatible models. Installs a helper, not a runtime adapter: run enable-capabilities.ps1 once per profile (Windows) and restart — it declares the levels in ~/.dsh/settings.yaml, with a disable script and a backup to revert.

DeepSeek Harness model selector upgrade: one-level model & reasoning-effort picker with 5 Liang tiers, custom model aliases, and peak/off-peak price countdown.

LLM compatibility auto-healer: detects 400 errors from incorrectly declared model capabilities (thinking/reasoning) on transit/gateway providers, then idempotently patches `llm-pi-ai` config and suppresses thinking on the next request — no restart needed. Full settings UI for every pi-ai compat field.

Share any DeepSeek Harness session as a self-contained read-only static webpage — timeline, collapsible tool calls & reasoning, default redaction, dual SHA-256 fingerprints, watermark. · 把 DSH 会话一键变成可分享的只读静态网页：单文件 HTML、离线可开、默认脱敏。

Configure model input capabilities and reasoning efforts for DeepSeek Harness

Model-facing accurate_vision tool: precise image spatial reasoning via a vision model

DeepSeek Harness plugin with two default model-and-reasoning shortcuts above the composer and optional Settings customization.

FriendliAI serverless LLM adapter for DeepSeek Harness — OpenAI-compatible chat completions with dynamic model discovery and reasoning support

Reasoning-mode routing plugin for V4 Flash on headless: w7 persona anchoring, first-turn tool anchoring and complexity-dispatched guidance.

Splits the composer model selector into a searchable, favorite-marking dropdown and a reasoning-effort slider, with Ctrl+P / Ctrl+T quick-switch shortcuts.

Surgical rewriting of the model-visible conversation history of a DSH session (surface replacement); edit reply and reasoning chain separately. 上下文手术刀：强制改写会话的模型可见历史，回复与思考链分开编辑。

A GUI subagent-model selector beside the main model seat: pick a model and reasoning effort per session, applied to every in-process subagent.