Plugins

Zhihu inside DeepSeek Harness: hot, recommendation and follow lanes plus in-site search, formatted question and column bodies, a per-item discussion that runs as a real Session, and one-key export to Obsidian through its REST API. Reads with your own signed-in cookie, kept in the credentials service.

Paste a Douyin video link (easy to copy from the Douyin PC app) and the plugin downloads the video, transcribes the speech offline into timed Chinese text, and imports it into Obsidian as a Markdown note. Built for DeepSeek Harness and for beginners building a personal knowledge base. Needs Python 3.10+ with playwright (it drives Edge), PyAV, sherpa-onnx and numpy installed, plus a SenseVoice model placed under %USERPROFILE%\.dsh\douyin-kb\models.

Daily chronicle for DeepSeek Harness: collects session events, closes the previous calendar day after midnight into a Markdown week file (LLM scoring adds a daily summary; rule scoring does not), and extracts durable facts — decisions, conventions, pitfalls — that merge across days and can be superseded. 10 tools, a read-only sidebar panel, rule export, and a standalone Obsidian vault; zero third-party dependencies.

Obsidian vault tools: 13 tools for listing, reading and searching notes, creating, appending and editing by line range, deleting to the vault's .trash folder, frontmatter CRUD, tag queries, image insertion, batch operations, and renaming notes while rewriting the wikilinks that point at them; writes are atomic and keep CRLF and BOM.

Career knowledge base vault with ten agent skills for setup, capture and import, audit, JD resumes, interview prep and review, plus Feishu interop.

Pulls the bookmarks of X, YouTube, Reddit, Bilibili, Zhihu and Xiaohongshu into one DeepSeek Harness panel: a daily three-item review page, filters by platform, folder and role, and one-click export to Obsidian, Reminders or Calendar. Collection runs through the opencli browser extension, so it needs Chrome with that extension installed and logged in.

Adds a knowledge-base sidebar backed by the separately installed local Obsidian Quote service: select registered vaults or folders, retrieve read-only note excerpts and cited context, extract rules, generate project convention briefs, and view shared citation statistics.

Loopback HTTP/SSE service that lets an Obsidian plugin drive DSH sessions and inject vault notes into the system prompt (path-based, re-read every turn, so edits show up next turn); needs the companion plugin running inside Obsidian.

Obsidian-compatible Markdown notes in the DSH sidebar with search, tags, backlinks, graph, audio notes with transcription, a paint canvas, OCR (needs tesseract), client-side encryption and notes_* agent tools; the release tarball and the npm package carry a prebuilt Linux x64 daemon, other platforms build it from source with Go.

Stores the agent's long-term memory as plain Markdown in an Obsidian vault — every memory is a .md file with YAML frontmatter that Obsidian indexes natively, so you can read, edit, link and back up it with tools you already use. Nothing is locked in a database. Ships a sidebar panel (tag filter, sort by recency or importance, edit in place), 7 agent tools, three-tier token-budgeted recall, BM25 ranking (no embeddings, no API key, works offline, CJK bigram tokenization), importance decay with archiving (moved to .archive/, never deleted), and optional GitHub Contents API sync. A note containing a secret is refused rather than redacted, because a redacted memory reads as complete while the important part is gone. Zero runtime dependencies; everything comes from the DSH host.

Loopback HTTP/SSE service that lets an Obsidian plugin drive DSH sessions (list, create, send) and inject up to ten vault notes into the system prompt; needs the companion Bridge plugin running inside Obsidian.

Work-review loop for an Obsidian-style vault: writes a day's Dream with a numbered T/G plan, executes only the items marked routine (read-only by default, local commits only with --live), folds seven days into an ISO-week routine, and shows every day and week in a read-only sidebar panel.

Project record system for DeepSeek Harness: a node-based progress board, a decision ledger, and a right-sidebar UI.

Codebase knowledge graphs for DeepSeek Harness: seven graphify_* Agent tools, a /graphify command, and a Graphify page build and query a project graph locally with AST parsing (no API key), plus path/explain/god-node lookups and wiki/HTML/SVG/GraphML/Obsidian exports. Requires the graphify CLI (graphifyy); only full document/PDF/image extraction needs an LLM key.

Plain click on an Obsidian vault file in the DSH conversation opens it in the right app: notes in the Obsidian app, configured types (for example .tex) in VS Code with their project folder as the workspace; Cmd/Ctrl-click keeps the in-app preview. The open_vault_file tool follows the same routing, and every open is confined to the configured vaultRoot.

Local English-to-Chinese simultaneous interpretation: on-device Qwen3-ASR bilingual captions with automatic language detection, crash-safe meeting records with minutes, and Obsidian export.

Brain-shaped relationship graph viewer for DeepSeek Harness with a Markdown knowledge-base pipeline that imports documents and builds the graph. The viewer ships demo data and needs no Python; rebuilding the graph from your own notes needs Node.js 20+ and Python 3.8+ and runs from the command line.

Registers two tools for DeepSeek Harness. `doc_ocr` turns scanned PDFs and images into structured text (per-page text, financial tables with a balance-sheet tie-out check, contract-clause hits) through a local RapidOCR engine or a Zhipu GLM-OCR cloud engine, and `doc_preprocess` turns a directory of raw files into an Obsidian-ready knowledge base through a seven-stage pipeline with backend routing and sha256-based incremental skipping. Numbers that fail the tie-out are reported, not invented. The local engine needs Python 3 with rapidocr-onnxruntime, pypdfium2 and Pillow (pip install -r requirements.txt); the GLM engine needs ZHIPU_API_KEY.

Turns a local Markdown vault into two folders the user picks in Settings: a knowledge folder DSH writes notes into, and a memory folder used as the agent's long-term memory with per-turn recall and end-of-turn capture. Pure Markdown with frontmatter, BM25 search, reversible archive, and path isolation that rejects traversal, absolute paths and symlink escapes.

Opens the Obsidian vault a DSH session is working in, on the first human message — so your sync plugin (Nutstore sync, Obsidian Sync, Obsidian Git, …) starts running and the files the agent edits actually get synced. Skips a vault that is already open, and opens a new window behind the current one instead of stealing focus. Also serializes writers per vault — a second conversation's write is refused before it runs and queued until the holder is done, so a sync client cannot splice two conversations' edits together. Windows only.

DSH 工作组件插件：在「工作组件」设置页统一挂载并管理控制其他工作软件的 MCP 服务器（Office / Blender / Unity / Figma / Photoshop / Chrome / Godot / Windows / 桌面控制 / Notion / Gmail / Cloudflare / GitHub / ComfyUI / FFmpeg / Obsidian），另含按游戏分卡的 GameBot（Minecraft / 文明 VI / 视觉）、会话控制、「通道」分组的 Telegr

Resident search index living in the host process: hot_find (fuzzy filename and path match, with frecency and git status), hot_fuzzy (content search that tolerates typos, including in Chinese), hot_grep (literal and regex content search), hot_plan (groups hits per file so a batch of files can be read in order) and hot_stats. One in-memory line index is built once and rebuilt on demand, so a query does not spawn a process.

Obsidian daily notes, tasks and Day Planner time blocks in the DSH sidebar: a Monday-first month grid with per-day word-count and task dots, a compact timeline for HH:mm time blocks, the selected day's diary excerpt and task list, and write-back of checkbox state, task text and times to the Markdown files; a first-run wizard detects the vault or accepts a path.

Multi-role parallel argumentation for DeepSeek Harness: runs the real Codex and Claude CLIs in parallel, has the DSH model (Judge) write a third-party verdict back into the conversation, plus Obsidian-style single-agent direct chat and per-role model config.