Plugins

Local-only project memory with zero bundled runtime dependencies or external services: JSON as the source of truth plus generated PROJECT_MEMORY.md, token-bounded session-frozen Hot Memory for prompt-prefix cache stability, cached cross-project BM25 recall, trusted session/turn provenance, explainable pre-write duplicate/conflict governance, editable importance/pin/tags/supersede/archive lifecycle, configurable top-level turn-end distillation, and a bilingual Web GUI for search, inspection, diagnostics, and live settings.

LSP action surface for DSH: diagnostics, formatting, completion, code actions, symbols, signature help, inlay hints and rename, all backed by real language servers.

A Rust/ratatui terminal client that speaks the DSH SDK JSON-RPC protocol directly and runs standalone or as a profile bundle.

Honcho memory plugin by Plastic Labs: injects the user profile, session summary and relevant conclusions at session start, uploads each turn to Honcho in the background with secrets redacted, and adds honcho_search / honcho_chat / honcho_remember tools plus a /honcho status command; configured through the ~/.honcho/config.json shared with the other Honcho integrations, against hosted honcho.dev or a self-hosted instance.

Zero-dependency toolkit: time / encoding / json / calculator / csv / regex / markdown / diff / stat / schema — ten deterministic tools in one install.

Four-source migration wizard: move Claude Code, Codex, OpenCode and Hermes sessions, memories, skills, instructions and slash commands into DSH (/move wizard with approval gate and idempotent move.json, resumable sessions).

Zero-dependency tool suite (calculator/csv/diff/encoding/json/markdown/regex/time).

DeepSeek Harness 用量与消耗插件（dsh-usage-plugin）—— 每次调用的 token 用量/缓存命中统计、峰谷计费、余额查询、CSV/JSON/PNG 导出，可经桌面端一键安装或命令行 dsh plugin add 安装。

DeepSeek Harness bundle registering one skill that turns long-form production (academic papers, industry analysis, business commentary) into a 9-role pipeline: T1 literature / T2 data / T3 case scouts running in parallel, T4 analyst, T5 writer, T6 critical companion, T7 auditor, T8 finalizer run by the coordinator, T9 peer reviewer — across 6 phases with triangular evidence verification, a 26-check M-gate (25 scripted checks plus one human checkpoint), a G0-G16 audit including a Chinese AI-trace gate, peer-review scoring with journal matching, and 4 human checkpoints. Installs with `dsh plugin add lunheng-article-pipeline` (package.json#dsh.bundle.patch plus a lib/index.js entry that registers the skill); ships role cards for T1-T9, templates and 36 zero-dependency verification scripts, with five-language READMEs.

MCP management page in DSH Settings: list, enable/disable, restart and delete MCP servers, edit them in form or JSON mode, expand a server to see the tools it registers, paste-import MCP configs from 14 tools (Claude Code, Cursor, Codex, VS Code and more) in one go, and bind a server to project directories so only sessions working inside them can see or call its tools.

AI image and video studio as a floating DSH panel, so creation runs alongside the conversation instead of replacing it: text-to-image, image-to-image and multi-image composition at 1K-4K across eight aspect ratios; text-to-video and image-to-video with first-frame control at 4-12 seconds; short-drama mode that imports a script (.txt/.md/.json), breaks it into storyboard shots for preview and batch generation; and a prompt-expert workspace. Zero runtime dependencies.

MCP server manager for DeepSeek Harness: add, edit, enable/disable, restart and delete MCP servers from Settings → MCP 管理, with JSON export/import, four mcp_manager_* model tools and an HTTP API; install via npx, npm or cross-platform scripts.

Mobile client plugin (cordis + dsh.plugin.json).

ROS2 debugging toolset and robot-state vision analysis for DeepSeek Harness: node/topic/service/action/interface/TF enumeration, whole-graph topology JSON, rosdep checks, approval-gated builds and custom message scaffolding, GUI screenshots and multimodal vision observation, plus headless RViz2 offscreen rendering (low-poly meshes, direct pixel read, GPU passthrough - motion rendering at 30Hz) with parallel VLM realtime analysis.

Skills & MCP manager in the DSH web settings: browse/enable/import/delete skills across project and user roots, and manage stdio/streamable-http MCP servers with a real connection (tools registered as mcp__<server>__<tool>, config in ~/.dsh/mcp.json).

Manage MCP servers (stdio / streamable-http) for dsh web; tools are reached only through four middleware tools. DSH MCP 管理器：Web 侧边栏「MCP」入口，面板按状态分级展示 MCP 服务器（运行中/连接中/已配置/失败），支持快速接入（stdio / streamable-http 表单）与粘贴 mcpServers JSON 导入。已连接服务器的工具一律经中间层工具（ws_mcp_

Smart Charts skill: turns CSV/TSV/TXT/XLSX/XLS/JSON data files into standalone interactive ECharts HTML - 32 chart types including offline maps, 3 themes, no CDN. A --profile data profile (per-column dtype, cardinality, missingness, samples, statistics, entity grain, insight signals) lets the agent read the data and pick the chart itself; each chart embeds a fact sheet (plot_stats + data_preview) whose numbers are the only quotable values. LLM transform code runs in a restricted sandbox (blacklist + AST whitelist + safe builtins); --doctor self-checks the Python environment. Fully offline, ECharts bundled.

Bridge Claude Code's .claude/ directories into DSH with zero migration: project and user-level skills, commands, rules (one system-reminder envelope), agent files, hooks (PreToolUse/PostToolUse/UserPromptSubmit/ SessionStart/SessionEnd) and .mcp.json servers, plus Claude Code plugin marketplaces. Ships /cc-plugin for full plugin/marketplace management (claude CLI first, self-implemented fallback), /cc-resume to list and import Claude Code sessions into DSH as native event-sourced logs, and /cc-export to write DSH skills back as Claude Code skills. Two-way Claude Code <-> DSH bridge, versioned and released as dsh-claude-compat.

Fine grained per tool permission rules for DSH at the tools/pre-execute gate, deny and ask lists in Claude Code rule syntax (`Bash(rm -rf:*), Read(_secrets_), mcp__server__tool`), works standalone without migrating.

Imports Claude Code setup into DeepSeek Harness from a native settings screen or CLI, with dry-run previews for skills, commands, agents, hooks, permission rules, and MCP config. Codex and OpenCode remain supported as secondary origins, with collision-safe apply and post-move checks.

Opens the official chat.deepseek.com page in the DSH main panel through the desktop app's native browser view, so the login survives restarts, without an iframe or request proxying. A fork of JaMoLpE88/dsh-desktop-chat with the changes listed in its NOTICE. The login token is kept in plaintext in session.json.

Patched build of dsh-pocket for DeepSeek Harness: the access password is embedded in the QR code so the phone signs in with one scan, a read-only /dsh-pocket/entry.json endpoint lets the app re-sync the current public URL and password from the LAN, and LAN / Tailscale / Cloudflare routes fail over automatically. This is a patched derivative of shaobeichen/dsh-pocket — install it instead of the upstream plugin, never alongside it.

Same-origin in mechanism with the two loudest lines in context compression. The code-skeleton gate follows the skeletonization approach of Headroom (Apache-2.0), whose published headline is 20% fewer tokens for coding agents and 60–95% fewer tokens for JSON, same answers. The estimator channel follows TokenPilot (arXiv:2606.17016), which reports up to 60% lower cost for long-session agents. Both figures are theirs, quoted as-is; this plugin ships no benchmark of its own and claims no reduction of its own. What it adds for DeepSeek Harness: choose a compression profile, set the Auto Compact trigger level and toggle code-skeleton compression from one settings section, with exact DeepSeek V4 tokenizer measurement, same-revision count verification, and fail-open behaviour that keeps the original tool results on unsupported models. Within a turn it supports skeletonized information compression with a low original-text retrieval rate; at the tail of a turn it supports a turn-tail intent summary implemented purely by content substitution; context content stays safe, and the whole process runs inside the official JSONL compression.

All-in-one local memory: JSONL files per scope (session, project, global, or a custom name) with per-scope dedup, five model tools (memory_save, memory_search, memory_list, memory_get, memory_forget), automatic recall and hard rules injected into the agent pre-step, and a settings panel. Retrieval combines lexical IDF with a local embedding backend, which the panel can install and manage itself.