Plugins
Browse, filter, and install DeepSeek-Harness plugins.
1434 plugins found
dsh-desktop-ball
han8421/dsh-desktop-ball
Windows-only desktop floating ball with a bilingual WebView2 quick-chat window, including in-window question and approval cards.
dsh-soul-evolve
xiachuyoufeng/dsh-soul-evolve
Persona state and slow evolution for DSH: five state dimensions (valence, arousal, energy, intimacy, focus) move after each turn and decay toward baseline over time; four scene types (work, tech, support, chat) modulate reply length and warmth without changing the state itself; the Soul panel keeps a persona card plus weighted evolution entries and injects them into the system prompt persona-first under a 4000-character budget. Ships a browser panel with a persona card editor and a live state view (the five metrics plus the last six events), appends a JSONL trace to disk, and seeds a usable default persona on first install.
bibao (bibao-brain)
huajikano/bibao
A simulated human-brain core for chat: global workspace, emotional appraisal, dual-system memory with nightly consolidation, and an experimental sensory interface; the QQ bridge needs a local snowluma client and is off by default.
dsh-local-trim
leonlee114514/dsh-local-trim
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.
dsh-mood
sata260/dsh-mood
Tracks an AI companion’s emotion and personality with a status panel, update tools, and chat stickers.
dsh-model-visibility
tzyceng/dsh-model-visibility
Manage which models appear in the composer model menu. A settings page lists every provider's models with a per-model switch, provider-level show/hide, collapsible groups and filters by kind; switched-off models leave the menu while the model in use in the current session stays. The list can be searched by name or ID and filtered by capability (multimodal, reasoning) and by minimum context window (128K / 256K / 512K / 1M). Rows show the model kind, and chat models also show reasoning support with the adapter's effort levels, their context window and input modalities. Labels merge manual overrides, the settings document, the adapter, the relay's GET /models response, a built-in pi-ai / models.dev catalog, name inference and vendor corrections in that priority order, with the source named on hover and dashed borders marking inferred values. Any model can be edited by hand (kind, context window, modalities, effort levels); those edits are stored outside the profile and can be written back into the model provider's settings. A hard re-fetch drops every cache and asks the adapter and all relays again. A switch, on by default, lets relay routes whose models declare text-only input accept images in memory, without editing configuration files. The hidden list is stored in the profile patch and mirrored to $DSH_HOME.
dsh-obsidian-notes
revive123456/dsh-obsidian-notes
Obsidian-style notes panel in the DSH right sidebar: a tree over a local folder, Markdown preview with line numbers, edit and save, Obsidian callouts, tables and wikilinks, screenshot paste into the vault, one-click insert of a note path into the chat input, and batch archive or delete for sessions and workspaces.
dsh-chatting-pet
flhy666/dsh-chatting-pet
Desktop pet for DSH with voice chat, personas, memory, Bilibili and QQ context, and command-gated computer control; enable DSH built-in voice input first to use voice chat. A fork of dsh-pet (MIT), not installable alongside the original.
dsh-deepseek-web
irerc/dsh-deepseek-web
Adds a browserless deepseek-web provider that drives chat.deepseek.com, with an account pool (rotation, cooldown, auto-failover), in-panel login (browser, or email + password with MFA) and token usage stats.
dsh-onebot-channel
danshuyue/dsh-onebot-channel
Connects QQ private chats and groups to DSH over OneBot 11, adding qq_send and other tools so the agent can reply, send stickers and wait for follow-ups.
dsh-artifact-linker
loyalchiiina/dsh-artifact-linker
Adds a resident system-prompt rule asking the assistant to report every file it produces as a full absolute path (a soft rule, not enforced), and replaces the official context menu on a file link with eight actions: open in the sidebar, open in a floating window, open with the default app, reveal in Explorer, save a copy elsewhere, copy the file, copy its path, and add it to the chat box, so a result reaches a friend in one click (copy-file and save-as are Windows only). Adapts DSH kernel 0.2.2-rc.1.
dsh-assist
a740411040/dsh-assist
Personas, memory and opt-in proactive chat that resumes after restart for DeepSeek Harness Desktop. 社区预发布插件:白饭鱼女儿、角色记忆与重启恢复主动聊天。
dsh-xiaod-pet
use3346631993/dsh-xiaod-pet
Anime whale-boy in a maid outfit as a desktop pet for the DeepSeek Harness web GUI: four expressions follow the turn (thinking / success / failure), free dragging with remembered position, click-to-cycle, a bottom-right standee that reports each turn's token usage and think latency with its own balance bubble, a chat wallpaper, a liquid-glass composer, a cute persona that calls you 主人 with 喵~ style particles, and a single switch in Settings to turn the whole look off.
dsh-novel-fish
gongfpp/dsh-novel-fish
Read local EPUB, TXT and Markdown novels below selected native DSH chat regions on explicit expansion, preserving real reasoning and tool output without sending novel text to the model. Requires DSH >=0.2.1-rc.2 <0.3.0-0.
dsh-weapp-loop
bushizhanghan/dsh-weapp-loop
WeChat Mini Program dev loop for DeepSeek-Harness: check IDE status, launch the DevTools project, trigger preview and get the QR code straight into the conversation, parse compile errors into file/line/column, and watch it all from a sidebar panel. Six agent tools over the official DevTools CLI — honest about needing a one-time scan-login, never fakes a preview. Requires the Windows WeChat DevTools CLI and a one-time QR login.
dsh-chat-mode
starp1ume/dsh-chat-mode
Adds a Chat/Work switch in the DSH Desktop titlebar: Chat embeds the DeepSeek web chat (chat.deepseek.com) in a native webview with no DSH API calls from the plugin and is the default on first launch, Work keeps the native agent UI, the page stays mounted across switches, and a confirm-first dialog can copy chat text into a work-mode draft without sending it.
dsh-auto-model-routing
prgrmrwy/dsh-auto-model-routing
Use Jev to adjust main-chat reasoning effort without switching its model, independently select and apply fresh sub-agent capability and effort, and inspect decisions with per-reply allocation feedback.
dsh-wx-archive
opencamel/dsh-wx-archive
A wx_archive tool that turns a WeChat Official Account article link into a permanent Markdown snapshot via the 2100laike archive service: images are re-hosted so they never break, snapshots survive article deletion, resubmits dedup-hit and return the latest version without spending quota, a content_kind field tells text articles from sticker (image-post) messages, and refresh forces a refetch with a PAT. Needs the third-party 2100laike service; anonymous use is limited to 8 archives per day per plugin instance and a registered PAT raises that to 100 per day.
dsh-assistant-profile
istarrydust/dsh-assistant-profile
Agent persona: gives DSH one global persona (name, language, persona prompt) that every new session carries at the front of its system prompt, plus a long-term memory you grow from any chat in one click; three built-in presets ship with it, and a Big fat fish persona is active right after install; chat sessions can also be deleted from the sidebar row menu.
dsh-remote-groups
seren-715/dsh-remote-groups
Android remote access and multi-agent group chats for DSH, based on ds-harness-remote and DeepSeek-Bot
dsh-pet-plugin
zuoyunlai/dsh-pet-plugin
A photo-real desk pet for the DSH window: five personas with five expression sprites each, whose dance, hair-flip and walk actions play as 24 fps VP9 alpha video clips, plus a single-shot stage clip that walks to the centre, dances and walks back without cutting; proactive lines with an optional edge-tts neural voice, ink / confetti / cloud / fog / bomb effects, a right-click menu and a /dsh-pet command, plus a small chat box that answers through a one-shot host model call without writing to your session. UI text ships in en, zh, es, pt and hi.
PTT-Studio
coldmagicchess/ptt-studio
Design-first PowerPoint builder: six pptx_* tools turn a deck.json design draft into an editable .pptx with AI backgrounds, cutout illustrations and decoration pieces, run layout and contrast checks, render slide previews, and review pages with a configurable vision model; chat cards and a settings page in the Web UI. Requires DSH Desktop 0.2.1-rc.2+ (pptx_render uses its bundled LibreOffice Kit) and an Agnes or ModelScope API key for image generation.
dsh-image-lite
yupaoa/dsh-image-lite
Image generation for DSH in two modes — an image_generate_lite tool the model calls from chat, and a right-sidebar studio for manual runs; up to five reference images, the OpenAI images API format, keys kept in the DSH credential service.
imgpost
hige6/imgpost
Let the DSH assistant send local, web, and AI-generated images directly to users in chat, with external vision support for text-only models. On Windows it runs through PowerShell.