DeepSeek Harness プラグイン — 140 ページ

无痕会话（incognito）插件：DSH Web 里的临时子 Agent——独立 preset 不继承父会话、全量工具集、窗口可拖动、关闭即焚毁、磁盘零痕迹

自带 macOS 菜单栏灵动岛的 DSH 插件 — 鲸鱼娘桌宠 + DSH 浅蓝胶囊托盘（blue-fantasy 配色），实时显示会话/工具/审批，点击展开可直接批准/拒绝

DeepSeek Harness bundle that registers jev_ask for TypeSafe Jev noul, choice, and score answers.

Jev-judged context compaction for DeepSeek Harness. Two layers: prune stale tool results (semantic instead of size-only), and replace spent tool-call ranges with a deterministic receipt instead of a model-written summary.

DeepSeek Harness plugin: choose whether Enter, Shift+Enter and Ctrl/Cmd+Enter send the message or insert a line break, from the Settings window's General page.

Kimi (Moonshot AI) official Formula tools for DeepSeek Harness: a ctx.web search provider backed by the Formula API chat loop, plus model-facing tools for the remaining official formulas (convert, rethink, random-choice, mew, memory, excel, date, base64, fetch).

Use a Kimi Code membership in DeepSeek Harness through a dedicated Kimi subscription model route

Kimi WebBridge for DeepSeek Harness: drive the user's real browser (navigate, click, fill, snapshot, screenshot, evaluate, network, upload, PDF) through the local Kimi WebBridge daemon

Kuikly integration and app-prototyping expert: four bundled skills over the create-kuikly-app CLI

LAN/remote access gateway for the DeepSeek Harness Web UI: 0.0.0.0 binding, crypto.randomUUID polyfill, token gate v3 (401 login page incl. loopback + index-401 fallback), privileged-fence + settings-persistence exemptions. Trusted networks only.

Lark/Feishu IM bot channel for DeepSeek Harness: chats drive agents, replies and approvals return as messages and cards

A standalone Lark/Feishu gateway and persistent cron bundle for DeepSeek Harness

Multi-tenant Feishu/Lark OAuth login and per-user session isolation for the DeepSeek Harness Web GUI

Regression-aware, zero-config lint feedback loop for DeepSeek Harness — model-visible lint_diagnostics / lint_workspace_errors / lint_fix tools (eslint / biome / ruff / golangci-lint / cargo clippy, auto-detected) plus an auto-injected post-edit findings section, closing the edit → lint → fix loop without replaying historical lint debt.

Academic literature search, citation and BibTeX tools for DeepSeek Harness and any agent (Crossref + OpenAlex + Semantic Scholar).

Live2D companion for DeepSeek Harness: 8-state AI status lamp, web widget, and always-on-top desktop pet with expressions, motions and bubbles

Live2D companion for DeepSeek Harness: render a Cubism model (e.g. the official Haru sample) in the web UI with state-driven expressions and mouse tracking. · DSH 桌宠的 Live2D 渲染插件

Bilibili-style live-chat danmaku overlay for DeepSeek Harness Web: preset packs (general/coding/casual), optional LLM-generated quips with smart/interval/tool-call wake modes, draggable control ball, per-session danmaku pools with history replay + archive/restore, heat bar, and a full settings card. / 给 DeepSeek Harness Web 对话区叠加 B 站风格弹幕：预设包、可选 LLM 吐槽（智能/间隔/工具调用唤醒）、拖球、分会话弹幕库（历史回放+归档/恢复）、热度条与完整设置卡。

Command Code Provider API LLM plugin for DeepSeek Harness

Inject custom HTTP headers (user-agent etc.) into DeepSeek Harness LLM provider requests, configured through cordis.yml or the Web UI settings surface.

Ollama Cloud models, OpenAI-compatible chat, and Web Search/Fetch for DeepSeek Harness

Best-of-3/5 orchestration and LLM-as-a-Verifier selection for DeepSeek Harness

DeepSeek vision bridge for the dsh web GUI: route image attachments to a vision model (pi-ai / llama.cpp Qwen3-VL) and continue the conversation with the text description on a text-only LLM (DeepSeek)

DSH web profile bundle: 黑盒/代审 sidebar workspace (P0–P7, per-finding SRC reports).