Plugins

Four native tools for image/video model discovery, contract inspection, pricing previews, approved generation and resumable task status through BeatAPI.

Connects Harness to the published BeatBrowser runtime with namespaced MCP tools and a TaskSpace Skill; requires the Chrome extension and preserves manual takeover.

ClickHouse query tools for DSH, targeting ClickHouse alone: seven tools that list databases, tables and columns, sample rows, search columns by name, show EXPLAIN plans, and run single read-only SQL statements behind a client-side guard that rejects writes, multi-statement SQL, cross-table queries and external table functions. Single-table by default: the query parameter is used as given, with no built-in multi-step business flow. Zero runtime dependencies; it tells the model to backtick non-ASCII aliases instead of failing with a bare syntax error on older servers.

Connects the Tavily hosted MCP server and registers the tools it publishes. The API key is read from the TAVILY_API_KEY credential and is not stored in plugin config.

Runs the local Codex CLI from DSH as four tools (codex_run, codex_resume, codex_queue, codex_probe), and ships a standalone loopback bridge with an HTTP API and a file mailbox.

Runs the local Claude Code CLI from DSH as three tools (claude_run, claude_resume, claude_probe) that parse --output-format json and return the session id for follow-ups, plus a standalone loopback bridge.

Drives the Claude desktop app from DSH over its claude:// deep links: open a Claude Code session with a prompt and folders, a Cowork task with files, or push a CLI session id back into the app. Also presses Ctrl+Enter for you (SendKeys) and reads the reply back from the local Claude Code transcripts under ~/.claude/projects/<cwd-slug>/<session>.jsonl — seven tools: desktop_send, desktop_continue, desktop_open, desktop_sessions, desktop_read, desktop_wait, desktop_probe.

Five native tools for Social Data discovery, pricing previews, approved reads, stored-result retrieval and source tables; requires a BeatAPI key for paid reads.

Brings McDonald's China official MCP (mcp.mcd.cn) into DSH. Paste a token you generated yourself and the plugin handshakes, calls tools/list, and registers every remote tool as a native DSH tool (mcd_ prefix) — upstream additions and removals need no code change, just a restart. Ships four always-on tools (mcd_status, mcd_test, mcd_tools, mcd_call) for self-checks and direct fallback calls, plus a settings page with a one-click connection test. The token is stored only on this machine in ~/.dsh/dsh-mcd-mcp.json (mode 600); it never enters the repository or the conversation context.

Gives a DSH agent five layers of persistent memory kept as plain Markdown files — persona, speaking-style rules, daily diary, session archive and open loops — rebuilds a system-prompt section from them within a minute of a file change, and adds three tools (memory_read, memory_write, memory_search) plus an importer for migrating an existing memory directory.

Splash and exit animations for DeepSeek Harness. Five hand-drawn styles (whale light-sweep, pulse, stardust, typewriter, radar rings) — or drop in your own video / GIF. A settings panel controls speed, particle count, accent colour, wording, mute and fit. The exit animation needs the bundled one-time desktop-shell patch (node tools/patch-asar.mjs --apply, reversible); without it closing only hides the window.

Adds a conversation view that maps which skills, MCP tools and subagents a session actually used, with an attribution tree, a call table and an unused inventory.

A collapsible column beside the conversation that organises conversations into a workspace/topic tree, opens the real sessions in place, and exposes the tree to the assistant through 9 model tools.

ai-code-quality (cq) tools for DeepSeek Harness (dsh): deterministic code-quality pipeline — doctor/scan/triage/baseline/diff/gate/waiver/spec-drift — aggregating ruff, mypy, radon, vulture, pytest-cov, import-linter and DeepSec. Read-only scanning plus a

DeepSeek Harness plugin for OpenProject — manage projects, work packages, time tracking, members, versions and attachments from chat. 36 Chinese-localized tools, zero runtime dependencies.

Author your own skills in the Harness Settings page: every entry is published into the skill registry, so the model sees one line and loads the body on demand.

Codex-style file-based long-term memory for DSH: two-stage LLM extraction, consolidated MEMORY.md handbook, prompt-injected memory summary, and search/read/list/note tools.

Manage MCP servers, skills, and tools for your DeepSeek Harness agent — true in-context state (loaded/truncated/evicted) and per-session switches for every capability.

DeepSeek Harness plugin: a port of the ZCode official android-emulator plugin — 23 Android build / emulator / ADB / UI-Automator MCP tools plus the android-dev skill.

DSH agent tools for local Ollama System One decision models: classify, answer yes/no, and score against a rubric in one batched request.

Checks installed DSH plugins and previews upgrade risks, with source locations, dependency views, review records and evidence limits. Includes a panel and agent tool.

Context Transparency Manager: the model context as a visible, editable, effectiveness-scored surface (self-contained dsh bundle plugin)

Relvato website monitoring for DeepSeek Harness: adds Relvato's hosted MCP server (58 tools) through dsh-mcp-client, with a Relvato API key.

Memex for DeepSeek Harness (dsh) — every day it records what dsh did (sessions, prompts, tools, files, commands) as a Markdown diary. · dsh 的起居注：每天自动整理 dsh 当天做了什么，存成日记。