Plugins
Browse, filter, and install DeepSeek-Harness plugins.
1640 plugins found
dsh-gatekeeper
huajikano/dsh-gatekeeper
Inspects the model's streamed reply before it is committed and either discards it or holds it for review when it reads as a refusal, so that text does not stay in the session context. Detection combines built-in phrase rules (an opening refusal, a capability denial, policy wording, first-person intent, refusals quoted in reasoning) with a plain-text word list and optional regular expressions. Interception can sit at the llm/stream waterfall, at the HTTP fetch layer before encryption (not TLS decryption, and it cannot tell which session a call belongs to), at both, or in a preview mode that shows a transient preview while the original chunks stay in memory until the whole reply has been checked. The same plugin also guards tool calls on tools/pre-execute: a command that deletes a drive, filesystem or user-profile root, formats or repartitions a volume, empties the Recycle Bin, fetches and executes a remote script, or writes to a raw device is denied, and a recursive delete outside build directories is recorded by default. Each tier can be set to deny, ask, record, or off, and the disk guard has its own switch. Both halves are heuristics: they can miss phrasings, the disk guard sees the arguments of a call rather than what a running script does, and they fail open.
dsh-status-box
nuke0519/dsh-status-box
macOS menu-bar status box for DSH: a whale status item shows running / waiting / completed counts and colours, its menu lists sessions grouped by workspace, a top-bar pill mirrors the counts inside the window, and the global hotkey ⌥⌘⇧D opens a single-field capture window that queues a task and sends it to the chosen workspace as a new session (queued while DSH is closed, delivered when it opens). A settings page adds an open-after-send switch. Needs macOS 11 or later, the Xcode command-line tools and the desktop app. The menu-bar part is a separate native helper that you build and install with menubar/install.sh from a copy of the repo: it compiles with clang, registers a login item (LaunchAgent) and listens on 127.0.0.1:59421, where the plugin posts session titles and status every few seconds.
reminder-dsh
kes1ke/reminder-dsh
Native Windows completion and approval notifications for DeepSeek Harness, with session titles, optional AI summaries, quick reply, and an opt-in taskbar completion badge. Toasts carry the original session route, subject to Windows activation. The badge requires PowerShell 7.
better-ego-browser
cyrus-xc/better-ego-browser
Windows-only derivative of the official ego lite DSH plugin (@ego-lite/dsh-plugin, citrolabs/ego-lite) that cannot be installed alongside it: connects DeepSeek Harness to the installed Windows ego lite app through authenticated loopback HTTP for task spaces, navigation, page reading, clicks, form input and screenshots.
dsh-subagent-switch
ashenseeker/dsh-subagent-switch
Picks each subagent model from the local catalog by task category, unless the caller already named one. Free and quota tiers are ranked together by per-model capability tags; a pricier model is used only when no cheap model is tagged as good at that kind of work. Only providers whose remaining balance is readable, or that are account-backed, enter the candidate pool; the rest stay listed but are not routed. A panel adds per-model manual priority above the automatic order plus allow/revoke controls.
dsh-explorer-open
ame-code/dsh-explorer-open
Adds a right-click command "通过 dsh 打开" (open with dsh) to the Windows Explorer right-click menu for a folder, a folder's empty background and any file: it launches DeepSeek Harness on that folder, creates a new session every time, and puts the clicked file into the new draft as an @ reference. Ships a packaged-COM menu for the Windows 11 modern menu plus a registry entry for the classic menu; first install asks once for a certificate trust (native/trust-cert.bat).
dsh-composer-extension
mulibaixi/dsh-composer-extension
Links source material and main-chat instructions, and coordinates translation and refinement tasks with the companion translation plugin. Linked submission needs submissions protocol 1: when the host lacks it, Windows Desktop installs a bundled host patch on first use (a full exit and reopen is required); other platforms need a host that already provides it.
dsh-siyuan-sender
baby7/dsh-siyuan-sender
Sends a Markdown file from the conversation into a self-hosted SiYuan notebook: a button on every delivered-file card and on the document-preview toolbar, or just ask the agent to send it, all through the SiYuan kernel HTTP API. Existing notes are never overwritten — a repeated title gets a timestamp suffix — and the API token stays in the host credential store rather than in settings files or the browser.
dsh-workspace-session-filter
shijiewang/dsh-workspace-session-filter
Four filter chips on their own row at the foot of the sidebar, each carrying its match count beside the label: running sessions, sessions waiting for your approval, plan review or answer, sessions with something unread (a turn that finished unseen, or newer activity the panel was not opened on), and sessions updated since local midnight. A chip opens a popover listing its matches grouped by workspace, and a row opens the session. The row ends with a DeepSeek service status dot (green operational, yellow partial degradation or maintenance, red open incident, grey when the feed is unreachable) that subscribes to the status Atom feed once a minute with an ETag conditional request and shows the current finding, the refresh time and the three newest incidents on hover; clicking it opens status.deepseek.com. A separate panel adds an all-sessions chip, a more menu for the last day, three days, week or thirty days, and a manage mode that searches sessions by title or path, multi-selects them, lists the Markdown digests it wrote to the session-digests folder of the DSH data directory in a resizable, full-screenable right-hand column that previews them in place, and either moves the selected session directories to sessions/.trash or deletes them permanently after writing the digest and asking for a second, explicit confirmation. The session you have open cannot be selected. Every panel row also carries a star (shown on hover, filled once starred) and the panel adds a favourites tab that lists only starred sessions, so a long list can be triaged without adding another chip to the sidebar foot.
dsh-handoff-session
blueocean3214/dsh-handoff-session
Pressure-aware session handoff for DeepSeek Harness: writes a handoff doc before a long session degrades, then continues in a fresh session in the same workspace. Subagents are snapshotted, still-running ones are interrupted, and unfinished tasks are handed to fresh subagents; a session without subagents simply gets the normal handoff. The handoff request itself is editable from the panel. A status pip and its label shade from green to blue to amber to red as pressure rises.
Cute-Partner (harness)
suzuran0y/cute-partner
Brings Harness conversations and tasks into a Live2D character on the Windows desktop, with drawer-based memory. The desktop app includes the whale girl 哦鲸鲸 as its default character. The modular design supports optional voice components for voice conversations and custom character voices.
dsh-task-shelf
xin505/dsh-task-shelf
Organizes task materials and supports session handoffs across workspaces.
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.
cece_legal_skills
rainbowzhu22-eng/cece_legal_skills
A DeepSeek Harness bundle adapted from whatcccup/cece_legal_skills (MIT) for local DOCX, PDF, TXT and Markdown redaction with human review, PDF masking, local history and optional Agent rule edits. Requires Python 3.9+ and first-run dependency downloads. Model access is needed only for rule generation; 1.3.0 is a development prerelease and Windows acceptance is pending. The upstream is a standalone offline tool/SkillHub skill, not a DSH plugin, so it cannot be installed alongside this bundle.
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.
CatDogTrace (dsh-plugin)
xiebaiyuan/catdogtrace
DogCatTrace attribution parody as a right-sidebar tab: draw three integer challenges, paste each model reply into one box at a time, read the 24-breed dog and cat ranking.
claude-on-dsh
huajikano/claude-on-dsh
Runs a project's Claude Code opening questions and plan, hands the task to DSH, and offers post-turn Claude review choices. Windows only, and requires a locally installed Claude Code CLI (set claudeCliPath if it is not auto-detected). New sessions handed to DSH are switched to danger-full-access. The first Claude run in a project folder sets hasTrustDialogAccepted for it in ~/.claude.json, and optional automatic Claude review is capped by a daily budget ($5 by default).
dsh-docx-format
romebody/dsh-docx-format
Normalizes every generated .docx to one Chinese official-document layout: XiaoBiaoSong title, FangSong body at a fixed 29pt leading with zero paragraph spacing, character-unit 2-character first-line indent, fully centered table cells with repeating header rows, and a centered footer page number. Long attachment entries are split into real paragraphs at the wrap point measured by Word, producing a 2 / 6 / 5 / 6 character indent ladder. It applies the layout only when the request mentions 公文 (official document) and writes no signature block unless asked.
dsh-subagent-switch
wld12314/dsh-subagent-switch
Picks each subagent model from the local catalog by task category, unless the caller already named one. Free and quota tiers are ranked together by per-model capability tags; a pricier model is used only when no cheap model is tagged as good at that kind of work. Only providers whose remaining balance is readable, or that are account-backed, enter the candidate pool; the rest stay listed but are not routed. A panel adds per-model manual priority above the automatic order plus allow/revoke controls.
dsh-ask-first
baiye0303/dsh-ask-first
Adds a button to the composer that restates your draft, then asks a few questions with clickable options — every question offers an "I don't know, you decide" escape — and only then rewrites the draft into a request an agent can start on. The whole flow can be finished without typing, and it never sends for you.
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.
academic-agent (academic-dsh-plugin)
kw78/academic-agent
Local-first academic assistant: event-sourced courses, tasks, plans, reminders, course materials and 23 DSH tools.
dsh-xiaozhushou
flamencoff/dsh-xiaozhushou
A local task board with secretary, CEO, and historian roles, plus a desktop whale-girl companion for DeepSeek Harness.
dsh-power-switch
hongdougao111/dsh-power-switch
Adds a power button to the session header with two actions, each requiring a second click to confirm. Restart terminates the Electron main process itself (taskkill /F without /T), so no crash dialog appears: the host then exits gracefully on its own once the IPC channel drops, and a detached helper relaunches the same executable after the single-instance lock is released, confirming the new instance by a real window handle. Shut down does the same without relaunching. Leftover GPU and renderer processes from the previous run are reaped best-effort. Windows desktop app.