Plugins

Connects DeepSeek Harness to Baidu Netdisk's official MCP service for file search, file metadata, and storage quota. Requires Baidu account authorization after installation.

Cross-session long-term memory backed by Supermemory: pre-step semantic and chunk recall with per-session deduplication, debounced transcript upload, background LLM fact extraction (add/update/delete) with bigram and semantic deduplication, memory_search / memory_save / memory_forget tools, global or per-workspace container scoping, and a Plugins page UI for API key storage, memory CRUD, recall preview and activity log. Requires a Supermemory API key (a free monthly quota is available), and conversation excerpts are uploaded to Supermemory's cloud.

Fold the sidebar plugin-panel list to three entries, drag them to reorder, and keep every other panel one click away in a More menu — order and pins persist in localStorage.

Computer cleanup for Mac and Windows: disk cleanup, storage cleanup, caches, leftovers, similar photos; scans read-only and moves files to the Trash on macOS or a quarantine folder on Windows, with undo.

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.

Fairy III (Zenless Zone Zero) H.D.D skin for DeepSeek Harness: boot video, an eye-core dock that can be dragged vertically so it never covers text, and time-of-day greeting voices written by a model every day (5 lines, each <= 20 characters) and synthesized locally with GPT-SoVITS, with a fallback chain of today -> latest -> bundled audio so there is always something to hear. Replies can be spoken: a trailing marker line takes priority over length heuristics, and long replies can be compressed into one sentence first. The reader hashes the utterance it just played (MD5) so each reply is voiced exactly once, and switching sessions stays silent. Personal fork of dsh-fairy-zzz with its own package, route and storage namespace, plus a rate-limited summarize endpoint, POST-only engine control and bounded TTS and greeting caches. Two diagnostic logs (client revision, decision trace) are written next to the plugin and stay on the local machine. A personal fork of AssassinVIX/fairy-zzz-plugin (upstream package dsh-fairy-zzz): install this fork or the upstream, not both - both mount the full-screen skin on the same DSH surfaces. Dynamic greetings and spoken summaries need a DeepSeek-compatible API key in ~/.dsh/fairy-tts.json (or DEEPSEEK_API_KEY); without it only the bundled greeting audio plays.

Corbeille des sessions archivées avec déplacement par lots, restauration et suppression définitive, ainsi que l’usage du stockage DSH et des espaces de travail, les types de fichiers et les vues des gros fichiers.

Barre d’état de la barre latérale DSH. Pendant une session, la ligne devient une carte verte avec une progression fondée sur l’estimation de la durée du tour (piste pointillée animée) et un pourcentage en direct à l’avant. Une fois l’estimation dépassée, elle affiche le temps écoulé mm:ss. À la fin, la ligne devient orange et reste allumée jusqu’à l’ouverture de la conversation. Fonctionne uniquement côté client, sans service supplémentaire, avec quelques clés localStorage.

Trois thèmes photo pour l’interface Web DSH : River Night (三江夜), Lingyun Dusk (凌云晚照) et Emei Mist (峨眉夜雾). Chaque photo de Leshan couvre la fenêtre et la colonne de conversation avec un voile lié à l’apparence. Un moteur dérive les 119 autres jetons alias --dsw-* à partir de six entrées de couleur par apparence. Alt+K change de thème et le choix est mémorisé dans localStorage.

Ajoute une liste de tâches dans l'en-tête de session : le bouton fait défiler les tâches inachevées et devient rouge, ambre ou vert selon l'échéance. Le panneau flottant permet d'ajouter, modifier et cocher les tâches avec date limite facultative, et de replier la section terminée. Persistance dans localStorage et respect du thème DSH.

Marketplace de plugins auto-hébergé pour DeepSeek Harness avec comptes, groupes, permissions, stockage d’artefacts et statistiques d’installation. Lancez le serveur avec Docker et installez ce bundle client dans DSH.

Lit NTQQ (QQ 9.9.x) local sous Windows/Linux et exporte les chats de groupe/privés en txt/csv/json/md/html/sqlite. 16 outils : qqnt_status, qqnt_doctor, qqnt_extract_key, qqnt_index, qqnt_list_groups, qqnt_list_chats, qqnt_search, qqnt_stats, qqnt_export_group, qqnt_export_chat, qqnt_export_batch, qqnt_jobs, qqnt_media, qqnt_storage, qqnt_ask, qqnt_recall (refuse l’ambiguïté ; nettoyage sur liste autorisée). UI 127.0.0.1:46728 : index SQLite incrémental, 6 vues, tâches, une colonne sur écran étroit, expéditeurs masqués, médias groupés, découpage quotidien, préréglages, ZIP AES-256 protégé par mot de passe. OCR/voix/embeddings s’installent à la demande dans le virtualenv du plugin. Requiert python3/sqlcipher3 ; Paramètres crée le virtualenv. Clé lue dans QQ ouvert et connecté.

Mémoire locale à long terme pour DeepSeek Harness avec stockage SQLite, extraction automatique, rappel sémantique, corrections, oubli fiable et dialogue de synthèse dans l’application.

Mémoire à long terme intersessions pour DeepSeek Harness : injecte les règles épinglées à chaque tour et rappelle le reste selon des mots-clés pondérés par IDF. Déduplication par fact-key, stockage JSON simple et édition/suppression des entrées dans les paramètres.

Permet de démarrer des chats sans choisir de dossier de projet, avec un répertoire de travail persistant par chat et le stockage de sessions natif de Harness.

Six thèmes matière pour l'interface Web DSH : Ink Jade, Terracotta, Aurora, Abyss, Sakura Mist et Paper. CSS uniquement : un moteur dérive 119 jetons alias --dsw-* à partir de 6 couleurs par apparence. Alt+K fait défiler les thèmes et le choix est mémorisé dans localStorage. Sans images ni route hôte.

Historique persistant des prompts du composer DSH : Haut/Bas rappelle les prompts envoyés, même après redémarrage ; avec du texte saisi, Haut filtre par préfixe. Les brouillons effacés ne sont pas enregistrés.

Vérifie un registre de conservation des aliments d’une cantine : quatre éléments obligatoires, minimum de 125 g, durée de stockage au froid de 48 heures et marqueurs de modèle laissés dans une cellule.

Questions en arbre pour DSH : cliquez sur « Ask in tree » après une sélection pour créer un nœud dans l’arbre latéral droit et poser la question en ligne dessous. Un autre nœud devient le parent des sélections suivantes. Alt-clic crée une racine ; Ctrl-clic ajoute un passage au même nœud ; double-clic revient au texte cité. Un point rouge/vert indique si le nœud est compris. Le panneau alterne entre liste et graphe horizontal. « Add to composer » ajoute la citation dans la saisie comme puce de référence sans créer de nœud. Les nœuds sont dans `localStorage` ; rien n’est injecté dans le prompt.

Liste de tâches dans la barre latérale droite plutôt que dans la conversation : ajouter, cocher, supprimer et réordonner les lignes par glisser-déposer. Les éléments restent uniquement dans le `localStorage` du navigateur et le plugin n’enregistre aucun outil destiné au modèle. Un bouton dans l’en-tête de la conversation ouvre le panneau.

Coffre local chiffré de credentials pour DeepSeek Harness : gestion dans la barre latérale, outil info_records pour recherches avant connexion, capture de chat facultative et mode manual-only par entrée. AES-256-GCM protégé par Windows DPAPI, macOS Keychain ou Linux Secret Service.

Déplace toute une installation DSH et vérifie chaque élément. Couches : declaration (réglages, credentials, plugins, sources link vendorizées), state (sessions, stockage, skills, pièces jointes) et code workspace facultatif. Ne réécrit que le frame header zstd des sessions, remappe les chemins explicitement ; 14 contrôles (I1–I13 + I6b) rendent PASS/FAIL/WARN. Pairing LAN transfère le bundle ; merge s’arrête avant écriture si conflit non résolu.

Thème Firefly « Spring Missive » pour l’interface Web DeepSeek Harness. Remplace 74 design tokens --dsw-alias-* dans les variantes claire et sombre, dessine une fenêtre printanière avec pétales et lucioles, place une illustration de chaque côté et préserve la lisibilité de la colonne centrale par un halo radial. Ajoute dans Settings des interrupteurs distincts pour le thème et les illustrations ; les deux choix persistent dans localStorage.

Minuteur pomodoro classique 25/5/15 pour DSH, liant chaque cycle à la tâche de session actuelle ou à un nom de projet saisi. Inclut statistiques, carte d’activité sur 365 jours, minuteur flottant déplaçable et pastille d’angle ; stockage JSON local et export CSV.