Skip to main content

Plugins

Browse, filter, and install DeepSeek-Harness plugins.

128 plugins found

X

dsh-vision-hub (bridge-preview)

xing666173/dsh-vision-hub

Inline preview for \[图片: path] image bridge markers: after the marker renders, the image shows in chat automatically, keeping long instructions out of the conversation.

43 months agoSessions & Messages
Z

dsh-opengis-skill

znlgis/dsh-opengis-skill

Bundles the 32 gis/ skills of opengis-skills (GDAL, QGIS, GeoServer, PostGIS, PROJ, PDAL, GRASS GIS, JTS, GeoPandas, Shapely, CesiumJS, OpenLayers, Leaflet, deck.gl and more) into the DSH skill catalog; only names and descriptions appear up front, and instruction bodies plus reference documents load on demand.

34 days agoSkillsMIT
E

dsh-thinking-guard

ethanwong-hk/dsh-thinking-guard

Circuit breaker for idle agent turns: watches agent/assistant-stream and cancels a turn that produces only reasoning — no text and no tool call — after 45s, after 80000 reasoning characters, or on a degenerate repetition pattern (exact repeat unit, sentence-template repeat, sparse repeat, or 6-gram density). On trip it injects a continue instruction that requires the next action to be a tool call, and ships 11 regression cases plus per-gate tuning.

32 months agoDev & Plugin ToolsMIT
K

dsh-input-list

konglong87/dsh-input-list

DSH prompt library and reusable snippets plugin: save, edit, delete, and append prompt templates, workflow instructions, and reusable text from a button beside the composer without sending.

32 months agoUI EnhancementsMIT
B

dsh-agent-instructions

bay-zeddie/dsh-agent-instructions

Panel to edit dsh's native AGENTS.md from the web settings page: every layer in the instruction chain is clickable, three activation scopes, and a visual budget meter.

33 days agoIntegrations & RemoteMIT
L

work-personal-secretary (work-personal-secretary)

liangl1985/work-personal-secretary

Installer and configuration base for a five-plugin DSH assistant suite: probes the Python, WPS and workspace environment, installs the sibling plugins shipped in the same repository, and generates the workspace baseline (instruction layer, memory seed, skills, settings); it also contributes its own settings section and its own host-side setup and check logic.

3last monthIntegrations & RemoteMITArchived
L

dsh-memory-delta

lpf20200901/dsh-memory-delta

Layered cross-session memory for DSH that pushes only what changed - the first round injects every active entry, later rounds push only added / updated / removed ones, and nothing at all is injected when nothing changed, with keys that merely repeat an entry id left out of the text so the baseline is about 26% smaller. Facts and decisions carry semantic keys (one active truth per key), the model may only write to an inbox that a human promotes, an entry can carry a review date that raises a one-time reminder when due, and search is relevance-ranked with journal lines down-weighted and Chinese matched by bigram. A sidebar tab tells the lifecycle story instead of listing files - entries are grouped by stage (awaiting confirmation / in use / archived) along four axes (topic - one level, human-assigned, type, tag, date), each group can be ticked for batch promote / withdraw / archive / restore / delete / assign topic, and topics can be renamed store-wide or searched on their own. The archived stage reads in three levels - the category says why an entry left (distilled into the docs / useless for good / not needed right now, the first two being defaults you can add to), below it the entry topic stays collapsed until you click it, and entries with no topic sit right under the category. The tab also shows the global and workspace instruction files plus the current injection size. A zero-dependency CLI ships with it, so the store is usable without DSH.

3last monthMemoryMIT
1

dsh-prompt-polish

1321928757/dsh-prompt-polish

Input-bar prompt optimizer: rewrites drafts with 6 strategies × 3 languages plus custom instructions, optional goal/todo/compaction-summary/tool-result and chat context, a confirm dialog, last-5 history and three-tier settings persistence.

3last monthUI EnhancementsMIT
H

dsh-custom-instructions

huanghai-lab/dsh-custom-instructions

Manage global AGENTS.md in DSH Web with Markdown preview, reusable templates, version history, import/export, failed-import rollback, and revision-conflict protection.

33 months agoUI EnhancementsApache-2.0
B

dsh-composer-history

baisama-cloud/dsh-composer-history

Command history for the DSH composer: press Up/Down to cycle through previously sent instructions, then restore the draft at the end of the history; works in the official desktop app and web-profile builds.

32 months agoUI EnhancementsMIT
Z

dsh-personalize

zephyr-vibe/dsh-personalize

Per-host personalization for DSH: custom instructions, local long-term memory, and reply-tone presets.

33 months agoMemoryMIT
O

dsh-gpt-adapter

oh-dsh-plugins/dsh-gpt-adapter

DeepSeek Harness plugin for GPT models and Codex instructions, skills, plugin resources, and MCP configuration.

22 days agoMCP & ConnectorsMIT
O

dsh-multi-window-context-management

overact/dsh-multi-window-context-management

Windowed handoffs on top of DSH native compaction: each compaction appends a bounded window message (goal, todos, recent user instructions, notes), with session-private notes, paged history recall and a /ctx switch; notes and history tools appear after the first rotation.

26 days agoSessions & MessagesMIT
M

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.

29 days agoWorkflow & AutomationMIT
A

dsh-agents-viewer

atrestandbed-bot/dsh-agents-viewer

View and edit the AGENTS.md / CLAUDE.md instruction chain in a native right-sidebar tab — global, shared, and every project directory down to the session working directory — with in-place save that creates missing files, a chat-header toggle, and an /agents command that works without the panel.

2yesterdayUI EnhancementsMIT
S

dsh-agents-md

sillykjs/dsh-agents-md

Adds an Agent docs tab to the conversation view: it lists the AGENTS.md and CLAUDE.md instruction files DSH actually loads for the current session (the ~/.dsh and ~/.agents roots, plus every directory from the project root to the working directory), previews each as Markdown with per-file sizes and the 64 KiB instruction budget, edits and saves them in place after a sha256 version check, and opens the workspace folder from the path button; a saved file takes effect on the next message.

212 days agoUI EnhancementsMIT
L

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.

2yesterdayModels & ProvidersMIT
F

dsh-rice-mode

frisk325/dsh-rice-mode

Three-position toggle (off / this turn / always-on) for an "unlimited mode" that injects a "do not conserve tokens" instruction into the system prompt; the "this turn" mode reverts automatically on the kernel agent/status idle event.

216 days agoDev & Plugin ToolsMIT
Z

dsh-plugins (dsh-presets-zh)

zhiqiangme/dsh-plugins

Four Chinese agent presets in one plugin, Standard, PTC, Minimal and Creation (standard-zh, ptc-zh, minimal-zh, cordis-zh), each keeping the matching built-in tool set; Standard, PTC and Creation translate known host instructions into Chinese while tool IDs, schemas and SDK declarations stay unchanged; a settings page enables each mode and edits its name, description and system prompt.

214 hours agoWorkflow & AutomationMIT
D

dsh-galaxy-view

dialling/dsh-galaxy-view

Draws workspaces and conversations as a 3D star map in the stock panel slot: eight layouts (spiral, rings, armillary sphere, solar system, double helix, irregular web, cubic lattice, radial tree), eight colour themes, a memory-library map of instruction files, skills and skill calls, star-to-star fly-throughs, and a 中/EN switch. Links are drawn only where a workspace owns a conversation or one conversation descends from another.

27 days agoUI EnhancementsMIT
A

dsh-load-agent-rules

acus9345/dsh-load-agent-rules

DeepSeek Harness plugin that loads the rule files other coding agents use (Cursor .mdc rules, Copilot instructions, Windsurf, Cline, Kiro, Roo, GEMINI.md) with their glob scoping: always-on rules up front, glob rules when matching files are touched, the r

223 days agoDev & Plugin ToolsMIT
G

dsh-extensions

gulagala001/dsh-extensions

Manage skills, MCP servers, instruction files, custom commands and hooks. Hooks run commands with your own account outside DSH's sandbox and approval; the plugin asks you to confirm when you add one.

29 hours agoMCP & ConnectorsMIT
S

dsh-plugin-skill-autoroute

soh4c4759/dsh-plugin-skill-autoroute

Run one deterministic skill-routing pass on every user instruction, so the skill catalog is consulted before the model answers instead of after it remembers. On the host it scores the live catalog from the instruction's keywords and the catalog metadata, with no LLM or embedding call, and appends a single producer-tagged message to the step: a ranked shortlist by default, the winning SKILL.md body once the score clears a threshold, or an instruction to call a named router skill exactly once. Unrelated instructions stay silent.

2last monthSkillsMIT
T

dsh-windows2macos

twinsearth/dsh-windows2macos

Run one instruction on every online Windows and macOS machine, across networks and regions: a dependency-free relay talks to each machine over outbound connections only (Tailscale, a public VPS behind TLS, or a tunnel), every machine runs the command in its own checkout, and results come back with a commit, a working-tree fingerprint and a command hash so an environment mismatch is reported as unverifiable instead of being mistaken for a different result. Sending work needs its own operator token, separate from the per-machine device token.

214 hours agoIntegrations & RemoteMIT