Plugin

Article Review page in DSH's native right sidebar: turns Markdown, Word, PowerPoint and PDF articles into sentence-addressable text, lets you comment on any sentence, and hands the open review to the agent as a composer draft. Six agent tools let the model read every comment, revise the article (in place for Markdown, as a sibling <name>.revised.md for binary formats), resolve the comments, and record recurring wording preferences into a long-term writing-style memory injected into later prompts. Extraction includes a dependency-free DOCX/PPTX reader; PDF text extraction runs locally through the host's Python + pypdf (or poppler's pdftotext) and is unavailable without one of them. No network calls, no telemetry.

Windows companion window for DSH Desktop: todos, reminders, live session messages, approval handling and quick input. The standalone window reuses the Electron binary at %USERPROFILE%\.dsh\electron\electron.exe (usually installed by dsh-pet) and fails to open when it is missing.

Save any URL (intranet GitLab raw, public GitHub, etc.) to a local file — text or binary — with a text preview and skip-on-exists.

Exposes a DSH Harness Desktop Web GUI to another machine's browser through a Cloudflare quick or named tunnel, an SSH reverse tunnel, or a custom command, and hands out a tokenised link to open it; a settings panel and CLI start, stop and print the link. Needs DSH Desktop 0.2.1-rc.2 or later and a cloudflared binary.

DSH host bundle for the ShyPie privacy-first workflow: extracts attached images and PDF/Office files locally before they enter the session log (requires poppler pdftotext/pdftoppm on PATH and a local OCR engine — Windows OCR on Windows, optionally tesseract; without them attachments are rejected, not forwarded); default filter mode keeps text local (optional local masking); blocks unmaskable media/BLOB; provides local privacy_extract, privacy_ocr, privacy_scan and privacy_rehydrate tools; denies read_image and binary/document reads outside allowPaths; and enforces a fail-closed llm/stream egress guard before DSH dispatches a model request (the default provider allowlist is the local Magpie gateway).

Audio tools for DSH: inspect audio/video files, transcode or extract the audio track from a video, trim and normalize, and transcribe speech with local whisper.cpp or an OpenAI-compatible API. Requires ffmpeg and ffprobe on PATH (not bundled); local transcription also needs a whisper-cli binary and a ggml model.

One-step local model onboarding plus a light harness for DSH. On load it probes loopback and private-network inference endpoints (llama.cpp, Ollama, vLLM, LM Studio), reads each model's context window and output budget from the backend, and registers the provider through settings, so a local model becomes selectable without hand-writing config. It then trims only that provider: masks the tool list through tools.restrict, drops named system-prompt sections and memory contexts, swaps the injected instruction block for a short one while preserving its first 80 and last 96 characters, and truncates tool-schema descriptions without touching structure or required fields. Cloud providers are returned as the same assembly object, untouched. The injected block is re-read from disk on every assembly, so editing the file takes effect on the next request with no reload. The panel in the plugin detail page is a live operations view: it shows whether the takeover is active and by which hook, how many characters the last assembly was trimmed from and to, and which providers are online (probed concurrently, with a real one-shot completion reporting latency and reply text per provider). If a local backend is not running, the panel offers a one-click launch: it probes the port first (never starting a duplicate), then starts the matching server (ollama serve, or a custom command you supply) detached and polls until the endpoint answers, clamping the wait to 3-600 seconds, reaping the child it just spawned if it never answers, and naming the paths it searched when it cannot find the binary. A stop button kills backends this plugin started (SIGTERM, then SIGKILL) and refuses to touch a server somebody else launched. When a model has just been downloaded and no backend exists yet, the panel can also scan the machine for .gguf files with hard caps (300 models, 20000 directory entries, Downloads only one level deep), start llama-server on the chosen one (skipping the .mmproj. vision projection and reusing a server that already answers), register the provider under the model id the backend reports, and run one real completion to prove it works. Before spawning anything it does a fit check that counts only the part of the file the GPU cannot hold: hostBytes = fileSize - VRAM, compared against max(total RAM - 2GB, free RAM). A file far larger than RAM is refused outright; a heavy host-side remainder is refused unless you confirm; the same remainder under mmap is only a warning, because the weights land in reclaimable page cache. A load-mode selector (auto / mmap / none / mlock / mmap+mlock) maps to llama.cpp --load-mode, and the default is whatever the engine picks. From 0.4.2 everything above is derived from the platform instead of assumed: home and data directories come from os.homedir() and the OS conventions (LOCALAPPDATA on Windows, ~/Library/Application Support on macOS, XDG_DATA_HOME elsewhere); backend binaries are looked up in the official install locations of each OS and can be pinned with OLLAMA_BIN, LLAMA_SERVER_BIN, VLLM_BIN or LMSTUDIO_BIN; model discovery walks DSH_LOCAL_MODELS_DIR, ~/models, ~/llama.cpp/models, ~/.cache/llama.cpp and the LM Studio data directory, with the WorkBuddy layout demoted to one candidate among them. VRAM is asked from nvidia-smi, then rocm-smi, then sysctl hw.memsize on Apple Silicon (flagged unified, so the same bytes are not charged twice), and any flip to a unified-memory budget is handled explicitly. Before building a command line the plugin runs llama-server --help once and keeps only the flags that build actually has, so older engines get --n-gpu-layers 99 and no KV quantization while newer ones get auto plus q8_0; anything it cannot probe is passed through rather than dropped. LM Studio is started with lms server start --port N rather than being lumped in with the engines that need a model argument, and vLLM gets its own --model example instead of llama.cpp syntax. Windows, macOS and Linux are all exercised in CI, including a windows-latest and a macos-latest job, because a cross-platform claim that has only ever run on one OS is not a claim. Two of those suites exist specifically to catch this machine's blind spots: one replays the whole suite with process.platform rewritten to each OS, and one deletes every advantage this development box has — no nvidia-smi, no rocm-smi, no sysctl, an empty PATH, a home directory with no models — and requires graceful degradation rather than a crash. That last one caught a real class of bug: several assertions hardcoded 'lms.exe', which cannot pass where the executable has no suffix. Measured on the reference machine: a 16.9GB 27.2B MoE with 14.5GB of VRAM holds a 285-691MB working set and sustains 76 tok/s after a 1000-2000 token warm-up, so file size alone is not a fit signal. From 0.4.3 the scanned models are shown as a table rather than a dropdown (one row per gguf, with its size and any fit warning), so you can see what is on disk before choosing. From 0.4.4 the delete button on the provider table removes the provider entry, not the weights file on disk. That distinction matters: what people actually want to clear out is the dead entry - one pointing at a port that never answered, one registered twice for the same port, one whose model file was deleted long ago - not a 15.8GB gguf. Deleting a config is undone by importing again; deleting a gguf is not. The removal goes through settings.mutate with an unset op on \['providers', id], because update merges layers: writing null there leaves a null provider key that the loader reads as a broken config rather than an absent entry, and only mutate/unset is implemented as Reflect.deleteProperty. The result is read back afterwards, because unset on a key that comes from a lower layer (the profile cordis.patch.yml or a home patch) does not delete it - it restores the inherited value, so the call succeeds while the entry stays. When the entry is still there the panel says so and points at the file to edit instead of reporting success, and a DSH whose settings service predates mutate() is told plainly rather than silently ignored. Deleting takes two clicks and clearing the confirmation is as easy as clicking another row; on success the online and baseURL caches are cleared and the tool surface is recomputed, since the provider list changed. The gguf itself is never touched by this plugin - the scanned-model table is display and selection only.VRAM reporting is driver-dependent. From 0.4.5 model discovery also reads Ollama's own storage: its blobs are named sha256-… with no extension, so extension matching could never list them. Weights are now accepted by extension OR by GGUF magic header (only dot-free files of at least 8MB, capped at 600 header reads per scan; obviously-not-weights extensions like .json/.md/.safetensors/.part are never opened). Ollama models are resolved from their manifests: the model layer is picked by mediaType application/vnd.ollama.image.model (never adapter/mmproj layers), the human-readable name is restored from the manifest path (…/qwen3/8b becomes qwen3:8b, …/latest stays bare), tags sharing one blob merge into a single row, and the metadata kv supplies name, quantization and context length. Adopting an Ollama model does not start a duplicate llama-server over the same weights: it confirms ollama serve answers on 11434 (launching it if not) and registers the provider under the tag the endpoint itself reports — the blob's sha256 name would never match anything, since 11434 lists qwen3:14b, not sha256-a8cc…. Two real-machine-only traps fixed on the way: existsSync('<file>/') returns true on Windows, so manifest-vs-directory branching now goes by parseable content instead of trailing-slash semantics; and the quantization label table had been copied from a stale ggml enum generation, labelling ggml_type 15 as Q5_K_S while ollama show reports Q4_K_M. Same-blob dedup is by (dev,ino), keeping whichever entry has a name. Verified on the reference machine: the scan goes from 1 row to 4 — qwen3:14b (8.64G) and qwen3:8b (4.87G) now appear with Q4_K_M badges — in 4ms; npm run verify exits 0 (launcher 57, portable 35, panel-host green including the linux/darwin simulated-platform runs); the new panel-host suite builds a real miniature Ollama store in tmpdir and points OLLAMA_MODELS at it, so results no longer depend on whether the test machine happens to have Ollama installed. From 0.4.6 the panel shows each model its real name instead of its file name, because some download tooling stores every weight under one generic file name, leaving the real name only inside the GGUF header. The scan now reads each file GGUF KV section - not a fixed-size head buffer, because tokenizer.ggml.tokens (hundreds of thousands of strings on a large MoE model) and the chat template push general.file_type past any window; and GGUF string arrays are interleaved (u64 length plus payload per element), so treating them as a length table walks the cursor off the rails and produces astronomically large reads. Reading and skipping are accounted separately: key names and types are always read, string values over 256KB and string arrays are only walked past without copying, non-string arrays over 64 elements are skipped outright, and a cursor past the file size stops the read, because an out-of-range position means the cursor is wrong rather than that a limit should be raised. The displayed name goes Ollama tag, then general.basename plus size_label, then general.name, then the file name, with the full string kept for the hover text since it usually carries internal build numbers; mmproj files are now identified by general.type as well as by name. A short read (truncated file) no longer feeds half a value into the parser. On the reference machine, reading one file header parses 44 keys in about 16ms, and a whole scan takes about 31ms. From 0.4.7 the smoke test no longer fails the first time and passes the second. The old code gave both request attempts one shared AbortController and one shared 30-second timer, so the total budget covered both rounds: a cold load of a large model burned it in round one, the abort left the signal permanently aborted, and the second round (the Ollama-style reasoning_effort: none retry) could not even be sent - which is exactly why the first click returned This operation was aborted and the second succeeded. Now each attempt gets its own controller and timer, a timeout retries once automatically (retryOnTimeout, on by default) because a first-request timeout almost always means the weights are still being read from disk, and non-timeout errors (refused connection, HTTP error) are not retried at all. Timeouts report themselves: the message says the wait exceeded smokeTimeout, whether a retry already happened, and points at the new smokeTimeout setting (seconds, default 60, clamped to 5-600) - port responding is not the same as tokens coming out, so it is a separate knob from launchTimeout. Unrelated: an example naming one specific model was dropped from the panel copy, since not every machine has that model downloaded, and the explanation now describes the mechanism instead. From 0.4.8 adopting a plain .gguf no longer registers an unrelated Ollama model. The port field used to follow the last click rather than the file: selecting an Ollama entry moved it to 11434, selecting a plain .gguf did not move it back, so a 11434 left over from an earlier click stayed. The server accepted it as given, probed the port, found Ollama answering and borrowed it (the skipped-launch log line about the port already answering is this path), and then the model id - derived from the file name - could not be matched against Ollama's /v1/models, where the fallback was to grab the first entry in the list. Embedding models tend to sort first, so the adopt registered a model the user never selected. Now the port follows the file: a plain .gguf landing on another engine's port (ollama 11434, vLLM 8000, LM Studio 1234) is moved to 8090 and the panel says so. Only ports belonging to a known engine are overridden - a hand-typed 9999 aimed at your own llama-server is left alone. The model-id fallback no longer guesses either: it takes the only entry when there is exactly one (a single-model backend may legitimately report a different id than the file name) and otherwise reports an error listing the ids it did see, rather than registering a wrong provider. From 0.4.9 the panel can no longer vanish. It used to be possible for the whole section to disappear from the plugin detail page while the host stayed up, the HTTP bridge kept answering, and no crash log appeared. The cause was a scope error: a helper was declared inside one component and referenced from another, which is valid JavaScript and passes a syntax check, so it only threw at render time - when the user pressed the scan button and React ran the state updater. The host wraps every slot entry in an error boundary whose fallback is an empty anchor, so the throw was contained silently: the panel was not uninstalled or crashed, it was replaced by an empty div. The helpers now live in the module scope both components share. The client half also gained its own test - a minimal React stand-in loads the real panel module, really calls the components, really runs the state updater at render time, and presses the scan button, so this class of error can no longer reach a release.

Matrix-style binary (0/1) digital rain for the Harness interface: columns spawn only while a task is running, fall off screen when it stops, and leave a blank canvas; frosted-glass token layer with adjustable digit and panel opacity.

Browser in-app per DSH: Chromium dedicato con profilo persistente affinché gli accessi Feishu o SSO resistano ai riavvii, controllato via CDP. Espone sette strumenti agent (status, open, eval, read, feishu, screenshot, tabs) e mostra la pagina in tempo reale in un pannello destro aperto automaticamente dalle chiamate. Converte anche documenti Feishu in Markdown scorrendo il contenuto virtuale. Richiede Chrome, Edge, Chromium o una build Chromium di Playwright installata localmente (non ne scarica); specifica il binario con DSH_IAB_CHROME.

Raggruppa i 207 strumenti di reverse engineering REA (backend Ghidra / Hopper / IDA) in 3 comandi e 6 voci di dominio in sola lettura: circa 5 KB di testo strumenti residente invece di 757 KB. Nessun engine incluso; rea-agents, Ghidra 12.1.x e JDK 21 si installano separatamente.

Strumento di ritaglio per il composer: blocca lo schermo, consente di trascinare un’area, annotarla e inserire il risultato nel campo del messaggio. Solo Windows; il selettore è un binario WinForms incluso (ricompilato dal codice C# incluso se manca) e usa il .NET Framework 4.x integrato.

Strumenti Agent che scaricano video Bilibili, li trascrivono con whisper.cpp locale e creano note Markdown strutturate con l’LLM DSH; PDF facoltativo. Solo Windows. Se mancanti, scaricano nel data directory yt-dlp.exe, build ffmpeg.exe da GitHub/gyan.dev, binario whisper.cpp x64 da GitHub e modello ~547 MB da hf-mirror. L’esportazione PDF avvia Edge/Chrome installato localmente. Due strumenti: video_to_notes e note_to_pdf. JS puro, nessuna API key aggiuntiva.

Workspace remoti per DeepSeek Harness: scegli un host SSH, monta via SFTP una directory server come Workspace e leggi/scrivi/modifica con i normali tool file. Il controllo remoto fa polling per aggiornare l’albero; anteprime immagine/binario leggono byte-range. `remote_exec`, `remote_search`, `remote_git` offrono shell/ricerca/query Git remote. Una patch opzionale instrada bash/pwsh dell’agente al server. Ogni scrittura/comando remoto segue sandbox e policy di approvazione della Session; il plugin gestisce gli host con password, chiave o SSH agent.

Trasforma screenshot già presenti nel repository plugin in un hero reel silenzioso in loop di 6 s: webm VP9 con alpha reale, GIF con matte, frame di copertina e manifest con hold per scena e sha256. La timeline è esatta: hold interi + transizioni = `durationMs`; l’ultima scena torna alla prima (t=0 = t=duration). Nessuna dipendenza runtime: PNG con node:zlib, compositore di schede SDF e ffmpeg come unico binario esterno, rilevato a runtime. Pannello per scegliere, riordinare, generare e copiare il Markdown README; CLI con contratto `storyboard.json`.

Integra il server MCP dell’app desktop pen.dev in una sessione DSH, così il modello può leggere e modificare il documento `.pen` aperto in Pen tramite get_app_state, read_skill, execute, get_style e strumenti browser. Cerca il binario del server nei percorsi d’installazione usuali e aggiunge uno skill pen-dev e un listener che trasforma gli errori di limite del piano pen.dev in un breve avviso bilingue. Non ufficiale, non include binari pen.dev.

Fissa una barra di avanzamento sempre in primo piano durante il turno dell’agent; la percentuale è stimata dalle chiamate agli strumenti e arriva al 100% al termine. Mostra la fascia tariffaria DeepSeek punta/fuori punta; le operazioni con molti token chiedono conferma solo in fascia di punta. Processo Electron separato, saltato se il binario Electron non è presente.

Apre mesh STL nell’anteprima documenti della barra laterale destra: parsing binario/ASCII nel browser, viewport WebGL e statistiche di dimensioni, volume, area superficiale e tenuta stagna.

Registra doc_ocr_sign_fix in DeepSeek Harness per macOS. Diagnostica e ripara Python incluso che non carica estensioni native onnxruntime/cv2/pypdfium2 per hardened runtime/library validation: aggiunge entitlement disable-library-validation, rifirma l’interprete e verifica gli import. Idempotente, salva prima il binario originale e non tocca mai il bundle esterno dell’app.

Solo macOS arm64; richiede dsh-pet installato (il binario incluso è precompilato e chiuso). Un interruttore sul desktop mostra o nasconde l’animale con un clic e può avviare DeepSeek Harness se è chiuso.

Apre mesh STL nell’anteprima documenti della barra laterale destra: analisi binaria/ASCII nel browser, viewport WebGL e statistiche di dimensioni, volume, area superficiale e tenuta stagna.

Provider computer-use DSH Cordis per il desktop ZTE Nde. Il postinstall di npm install scarica nel bin/ del pacchetto il binario musl creato da computer-use-linux-zte-nde (v0.7.7-nde.3). resolveCommand() risolve il percorso cordis da import.meta.url, così il mount funziona indipendentemente dal cwd di dsh web. Prenota lo slot computer-use condiviso tramite ctx.computerUse.register ed espone tool MCP con prefisso mcp__computer-use-linux-nde-mcp__* (screenshot, list_windows, click, type_text, doctor ecc.). Due correzioni Nde upstream: fallback KWin caption+coords per selezionare finestre e SnapshotCache limita AT-SPI a un risveglio per (app, pid), evitando di disturbare Chrome ripetutamente.

Richiede piano obbligatorio di 7 passi per richiesta con criterio binario; worker subagent eseguono, verificatori indipendenti senza memoria restituiscono PASS/FAIL con prove.

Plugin host DSH che controlla dalla chat il gestore locale delle versioni runtime ployenv: elenco/installazione/uso dei runtime, binario ployenv incluso e lista dei percorsi inserita nelle shell di sessione.

Workbench di reverse engineering: collega IDA Pro locale a DeepSeek Harness tramite `@deepseek-ai/dsh-mcp-client` ed espone gli strumenti `ida-pro-mcp` (decompile, xref, rename, comment) come `mcp__ida__*`. Sceglie automaticamente `idalib` headless (analisi del binario nel processo, senza GUI IDA) o modalità GUI (RPC del plugin IDA sulla porta 13337). Aggiunge `re_check`, `re_open`, `re_status`; il server MCP incluso è un proxy stdio senza dipendenze utilizzabile da qualsiasi client MCP.