Plugins

Hands secrets and PII to the agent without them entering its context, with checksum-validated extraction and masked content classification.

Monitor CPU, memory, and UI responsiveness in DSH on Windows x64, inspect processes and sampled backend plugin resource usage, and copy diagnostic reports.

Windows commit-charge watchdog for DSH Desktop: samples committed memory every 5 seconds, attributes it to the DSH process tree and the largest consumers, blocks automatic restart when DSH is not the main committer, and adds read-only pagefile advice, a 7-day crash timeline and an event export; automatic restart and crash resume are both off by default.

Session-scoped research evidence ledger with sources, linked claims, open questions, and explicit user review; provides a tool, slash command, and conversation panel, with selected-claim Markdown export that omits excerpts, review notes, and local paths by default; does not fetch sources or verify their truth.

Session-scoped delivery receipts for local artifacts: compare current file hashes with a recorded baseline, keep agent claims separate from explicit user acceptance, flag changed or unavailable files, and export a Markdown receipt through a tool, slash command, and conversation panel.

Personal project assistant for bounded project context and memory, explicit task handoffs through an existing WeChat bridge, reminders, and source-linked summaries. The WeChat path needs the separate pan17/dsh-wechat bridge and your own scan-code binding.

Read-only Automations tab in the conversation view. It aggregates what other plugins are scheduled to run: last run time, next run time and a rough token-cost estimate, plus today's total spend and remaining balance. It owns no tasks and starts nothing.

Douyin patent and trademark complaint workflow for DeepSeek Harness: checks user-supplied links, materials and organization identity in a dedicated browser, records each submission state, and queries platform history read-only; never discovers targets or submits without explicit authorization.

Artifacts-like documents in the sidebar: the agent writes a self-contained HTML document, you comment on any part, send the comments back for the next version, or edit text and formatting yourself. Every change is a new version; local only. A local-only fork of tdoc (https://github.com/tornado-doc/tdoc), rewritten as a DSH plugin.

Turns a plugin repository's existing screenshots into a 6-second silent looping hero reel: a VP9 webm that keeps real alpha, a matted gif, and a cover frame, plus a manifest with per-scene holds and sha256s. The timeline is exact by construction (integer holds plus transitions sum to durationMs, and the last scene transitions back into the first so t=0 equals t=duration). Zero runtime dependencies: PNG decode/encode on node:zlib, an SDF card compositor, and ffmpeg as the only external binary, discovered at runtime. Ships a conversation-view panel (pick, reorder, generate, copy the README markdown) and a CLI with a storyboard.json contract.

A pixel whale girl living in your computer. The phone is the screen and the microphone; the brain is a DeepSeek Harness session on your own machine. Call the wake word and talk — the pet answers out loud and acts out the reply while it is still being generated. No accounts, no cloud transcript: pair a 30 MB Android app over LAN or your own tunnel. ASR uses the phone speech service, TTS defaults to Edge voice and can be swapped for local.

Local GPT-SoVITS voice for DeepSeek Harness: boots the engine with DSH and kills it on exit, speaks only the line marked for voice, and ships a default voice, so you only need to point runtimeDir at a local GPT-SoVITS install.

Dictation that ends up in the repository instead of the chat log: hold the button (or import a WAV), and the plugin archives voice-memos/<id>/ with audio.wav, a rendered waveform.png, memo.md (frontmatter, per-segment timestamps, ready-to-paste quote blocks) and meta.json. Audio is resampled to 16 kHz mono in the browser before it is sent. Transcription stays local — the harness speechToText service with silence-split windows, or a command worker such as faster-whisper — and when no engine is available the memo is still filed with an explicit status and reason rather than a fabricated transcript. Ships a conversation-view 口述 tab with click-to-seek playback and per-segment quote copy, three agent tools (voice_memo_save / voice_memo_list / voice_memo_read) and HTTP routes. Zero runtime dependencies, 52 node:test cases, real screenshots in docs/.

Video studio for DeepSeek Harness on Android: storyboard editing, multi-provider image/video generation with automatic failover, FFmpeg MP4 export, prompt optimization, and a mobile-first workbench.

Prove that a local image edit left every pixel outside its mask untouched. Original, generated and mask PNGs go in; a greppable `MASK_PIXEL_LOCK=PASS|FAIL|EMPTY` verdict, a difference heatmap, per-channel statistics, a bounding box and the largest changed pixels come out, plus a self-contained HTML report and `report.json` with a SHA-256 file hash, a canonical protected-pixel digest and a proof digest binding both to the policy. Supports edit and lock mask polarities, a mask threshold, tolerance and alpha-ignoring; PNG color types 0/2/3/4/6, bit depths 1-16, all five filters and Adam7. Zero dependencies, local only — no model call, no network, no upload.

Fills in the vision capability that DSH's "fetch models" step drops, so image-capable models stop being declared text-only, and marks them in the model name.

Browser tools for DSH that POST to the local OpenCLI daemon instead of spawning the opencli CLI per command, avoiding its cold start: oc_status, oc_exec, oc_navigate, oc_snapshot and oc_tabs, with one persistent automation tab per session.

Archives DSH conversation output into SiYuan notes: a single siyuan_archive call files content under `<archive path>/<year-month>/<type>/<date>_<time>_<topic>_<content>` with a metadata table and auto-maintains an index document at the notebook root; 6 siyuan_* host tools (list notebooks, full-text search, read document, archive, delete block, remove document) and a Web settings page for base URL with connection test, API token in the host credential store, archive target and title template, plus notebook/document-level read/write/delete permissions with default deny and child levels able to only tighten.

Search BOOTH for VRChat avatar assets — clothing, shaders, base models, accessories, textures, animations and props — from a dedicated BOOTH tab beside the conversation or via two read-only tools, with Chinese keywords translated to Japanese before the request.

Windows desktop control for DeepSeek Harness agents: window and UI Automation automation, local Windows OCR, direct browser reads over the Chrome DevTools Protocol (DOM text, JavaScript, selector clicks), humanised mouse and keyboard input, browser tab management, and a live control panel. It probes the machine on first run, writes its own config, and degrades gracefully when a capability is missing. Windows 10/11 only.

Mounts the pen.dev desktop app MCP server in a DSH session so the model can read and edit the .pen document open in Pen through the get_app_state, read_skill, execute, get_style and browser tools; finds the server binary by probing the usual install paths, and adds a pen-dev skill plus a listener that turns pen.dev plan-limit failures into a short bilingual notice. Unofficial, and ships no pen.dev binaries.

Local ComfyUI image generation for DeepSeek Harness — discovers or starts the ComfyUI installed on the machine, generates a text-to-image result from a chat request, and attaches it to the conversation.

Enter-plan-mode tool: gives the model a callable `dsh_enter_plan_mode_enter_plan_mode` that switches the current session into plan mode, pairing with the built-in `exit_plan_mode` for a plan-then-review loop.

The composer predicts what you are about to type. Inline ghost text sits on the caret, measured against the editor's own font and line box, and Tab turns it into real text (Esc ignores it). The prediction is one short call through the session's own model router, seeded with your draft and the last few turns of the conversation, so an empty composer also gets a next-prompt guess written in the user's voice; suggestions that merely restate the assistant are dropped. Host side is one HTTP route with sub-path endpoints (loopback-only, 64 KB body cap, 2-minute result cache, 20-second miss cache, one in-flight call per session); client side is one 0-height overlay registered on conversation.input.overlay, with insertion guarded by the composer's draftRev CAS and a setDraft fallback. IME composition and slash commands are skipped. 16 node:test cases, zero runtime dependencies, real screenshots in docs/.