Plugins

Writes local JSONL summaries of per-turn tool counts and coarse verification signals without storing prompts, tool arguments, or result text.

One memory layer for every AI tool and agent: Context Bundle injection, prompt-time recall, MCP tools, and turn-end DSH thread capture.

Search and insert workspace file and directory paths with `@`, plus a `/h` menu for reusing prompts from the current session.

Optimize a raw instruction into a professional prompt through the harness llm service — four-section (Role / Task / Context / Format) or a headerless plain-text style. Context-aware by default: the recent conversation is injected as background and may enrich the ## Context section. Long outputs resume from the truncation point with auto-expanded budgets; validated results are cached (identical requests cost zero model calls); a unified call budget bounds worst-case cost. Composer ✨ one-click optimize, undo, cancel and a token-cost hint; prompt_optimize tool (optimize + iterate), auto-optimize hook, auto-detected zh/en role document.

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).

DSH plugin: edit the system prompt (deployment persona) from the Settings page, with live preview.

Restart DSH: configurable restart method (Node native / legacy PowerShell), post-restart continue prompt, optional watchdog auto-relaunch.

Auditable vision and cross-platform Computer Use runtime for DeepSeek Harness with source-preserving evidence.

Windows desktop launcher for the DSH Web UI: desktop shortcut, silent start, system tray with open/full-exit (stop service, close app window, quit tray), PWA app window with focus-instead-of-relaunch, install prompt, and task-completion notifications.

Obsidian-style [[wikilink]] mentions for the DeepSeek Harness web GUI: fuzzy-search note titles and attach their contents to the prompt

Runtime-switchable model output styles with Claude Code outputStyles parity, plus the output.render.* presentation protocol: a /style command, per-session persistence, systemPrompt injection, six built-in styles, a web picker, and a renderer registry with per-session/per-tool rules and /export.

Local, per-session LIFO prompt stash for temporarily setting aside unfinished composer text and safely restoring it later.

Edit user and built-in system-prompt sections with live preview.

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.

Chinese localization for DeepSeek Harness: translates all system prompts, tool descriptions, and runtime contexts into Chinese at runtime via the system-prompt/assemble waterfall (1788 entries, zero source modification, uninstall restores English).

One-click ✨ polish button for the composer: rewrites prompts and drafts with WorkBuddy humanizer-style rules (kills AI-flavored phrasing, varies rhythm, mode-aware for instructions/questions/prose) via free Zhipu GLM-4.5-Flash or any OpenAI-compatible API, falling back to offline local rules.

Cost tracking priced from an online catalogue (models.dev + OpenRouter, 8000+ models, each call priced at its own instant): a per-turn cost line, a Cost tab attributing spend to tool output / model output / system prompt / commands, budget, monthly forecast, ~166 live-rate currencies, and a bill_stats agent tool.

Context-aware next-message suggestions in the DSH composer, with repeated Trigger cycling and explicit user review before sending.

DeepSeek Harness (DSH) plugin: a "Thinking Language" selector under Settings → General that tells the model which language to reason in (chain of thought) via a system-prompt section. Zero extra calls, zero latency, 12 languages.

Security-audit methodology skill pack plus the plugin_vet supply-chain gate: eight agent skills (secret scan, dependency audit, supply-chain review, prompt-injection review, audit orchestration, threat modeling, vuln intel, incident response) in Chinese and English editions, with an npm provider bundle that mounts the skills and registers the automated plugin_vet pre-install scanner.

User preference memory for DeepSeek Harness: persisted user profile with memory_get/memory_update tools and system-prompt injection

A model-native meta-agent plugin for DeepSeek Harness: the model synthesizes task-specific workflows at runtime from five agentic workflow patterns (prompt chaining, parallel workers, router, supervisor, evaluation loop) and coordinates tools and subagent

Scans skills and MCP configs for prompt injection, homoglyphs, hidden Unicode, dangerous shell, and credential leaks.

Per-file-type multimodal chains: preset processing models per wildcard convert image/video/audio files into prompt tokens before they reach the text-only session model, with per-preset fallback chains and a Multimodal settings page.