Plugins
Browse, filter, and install DeepSeek-Harness plugins.
96 plugins found
data-asset-inspector (data-visualization)
liuhange789/data-asset-inspector
Data visualization plugin that generates self-contained HTML reports with five chart types (bar/comparison/pie/radar/flowchart) and two-column layout.
data-asset-inspector (data-sensitivity-classification)
liuhange789/data-asset-inspector
Data sensitivity classification plugin that auto-identifies sensitive fields and classifies them into high/medium/low/public levels.
data-asset-inspector (data-cleaning)
liuhange789/data-asset-inspector
Data cleaning plugin for deduplication, format standardization and anomaly detection.
data-asset-inspector (data-asset-orchestration)
liuhange789/data-asset-inspector
Data asset orchestration engine that chains suite tools into preset pipelines (masking, cleaning, inventory, packaging, attestation, AI dataset health check, government inspection, circulation assessment), with incremental and scheduled processing.
data-asset-inspector (ai-dataset-inspector)
liuhange789/data-asset-inspector
AI dataset quality inspector for annotation consistency, class distribution bias and data leakage detection.
data-asset-inspector (data-masking)
liuhange789/data-asset-inspector
Automated data masking tool based on national standards, automatically identifying sensitive fields like ID cards and phone numbers.
data-asset-inspector (data-lineage)
liuhange789/data-asset-inspector
Data lineage tracking tool, mapping the full lifecycle of data to meet source legality requirements.
data-asset-inspector (data-asset-quality-score)
liuhange789/data-asset-inspector
Quantitative data quality scoring based on GB/T 36344-2018, providing auditable quality reports for data asset capitalization.
dsh-vaultwarden
buvkb/dsh-vaultwarden
Reads and writes a self-hosted Bitwarden/Vaultwarden vault from any session: eleven bitwarden_* tools, real-time WebSocket sync with a polling fallback, a settings panel with readonly, ask and auto write modes, and optional local masking — with masking on, passwords surface as stable placeholders unless revealed one field at a time, and on-disk session logs are swept for bitwarden secrets. Masking and log sweeping are separate switches.
dsh-sensenova-pool
phoeky/dsh-sensenova-pool
Brings SenseNova (日日新) models into dsh with a per-key rotation pool: paste any number of API keys and requests rotate across them, a 429 rate-limit immediately retries on the next key with per-key exponential-backoff cooldown, local RPM pre-throttling avoids the limit instead of reacting to it, and an exhausted pool answers 503 with Retry-After so a 429 never reaches the caller. Ships a built-in model catalogue that works offline and refreshes from /v1/models, plus a settings card with masked keys, per-key probing and removal.
dsh-theme-songgrid
yefengliu1/dsh-theme-songgrid
Song-dynasty-inspired theme for the DSH web UI: a light "rice paper" and a dark "night ink" variant, Ru celadon carrying interactive states while kiln copper is reserved for blockquotes, plus a 4pt spacing scale and a 46-glyph mask icon set.
dsh-medical-anonymizer
wld007/dsh-medical-anonymizer
Batch pseudonymization for medical text, CSV, JSON, and XLSX in DeepSeek Harness. Deterministic identifier rules run first; an optional OpenAI-compatible model pass adds discovery and verification. Exact-span replacement preserves file structure, tokens stay stable per value, and each batch produces an audit report plus a re-identification vault. DICOM images get header-only de-identification based on PS3.15 Annex E, with pixel data left byte-identical.
dsh-notes-plugin (dsh-notes-plugin)
ppawnsir/dsh-notes-plugin
Local Markdown notes as governable working memory for DeepSeek Harness agents: dual-mode source/WYSIWYG editor with images, convention or reference notes auto-inject into the system prompt with sensitive values masked, plus snapshot version history, recycle bin, undoable archiving, AI reorganization, backlinks and full-text search. Compatible with DSH 0.1.7-0.2.x.
dsh-wallpaper
rexcue/dsh-wallpaper
Wallpaper skin for the DSH Web UI: image background with opacity, mask and blur controls.
dsh-tool-drawer
kmyy350234-dz/dsh-tool-drawer
On-demand tool groups for DeepSeek Harness, with optional role-scoped tool surfaces. Role masks are off by default; Settings can enable or disable review/build/research, and add custom roles with a group allow-list. The role key comes from the host's session label, subagent marker or preset, so an orchestrator's child agents each see only their own allow-list. Also auto-groups tool families by name prefix, lets the agent classify ungrouped tools once, restores a hidden tool when it is actually called, and reports per-session visibility health.
dsh-document-redactor
rainbowzhu22-eng/dsh-document-redactor
A DeepSeek Harness document-redaction workbench adapted from whatcccup/cece_legal_skills (MIT). Processes DOCX, PDF, TXT and Markdown locally, with human review, PDF masking and OCR-line review, case-consistent placeholders, Word restoration, history and optional Agent rule previews and rollback. Requires Python 3.9+ and first-run dependency downloads; model access is needed only for rule generation. Version 1.4.0 is a prerelease verified on macOS Web and DSH desktop; new Windows setup and OCR paths await on-device revalidation. The upstream is a standalone offline tool/SkillHub skill, not a DSH plugin, so it cannot be installed alongside this bundle.
dsh-privacy-gateway
xbzbing/dsh-privacy-gateway
Redacts secrets, cookies and personal data in every model request at the host process undici dispatcher and restores the placeholders in the reply, so one interception covers every provider, adapters that plugins add included; per-kind actions mask / warn / block, an optional same-shape fake key, records written to disk by the day and a statistics card.
dsh-gpt-image-playground
southcul/dsh-gpt-image-playground
Image generation and editing for GPT Image, Gemini (Nano Banana) and Grok, with official, OpenAI-compatible, api.toolify-ai.com and custom HTTP providers. Text-to-image, reference-image editing, canvas annotation, mask editing and Agent multi-turn chat, with images and history kept on your machine. The workbench UI is the MIT-licensed GPT Image Playground v0.9.0 build by CookSleep, bundled unchanged; that upstream is a standalone web app, not a DSH plugin.
EVA-Inspired-Theme
whatcannotbesaid/eva-inspired-theme
A Neon Genesis Evangelion-inspired skin for DeepSeek Harness: Title Card (light) and Central Dogma (dark) modes over one removable override layer, 55 measured color tokens, a wallpaper paint layer whose two surface alphas are solved against a WCAG contrast budget, a boot mask and a brand mark. It also rewrites the sidebar identity line, replaces the new-session heading, hides six official entries, and reshapes the composer control row. Unofficial fan work, non-commercial: the background pictures are fetched from third-party URLs at paint time, not shipped in the repository.
dsh-jev-advisor
gordan-code/dsh-jev-advisor
Adds human-facing TypeSafe Jev advice to DSH multiple-choice prompts, on demand by default with optional automatic requests, recommendations, probabilities, confidence, close-call warnings, copyable suggestions, one-click adoption, profile-scoped card layouts, temporary decision preferences, optional request previews with common-credential masking, and local health checks with sanitized diagnostic exports.
stalegreen
pavangupta352/stalegreen
DeepSeek Harness plugin that keeps the agent's green claims honest: verification runs become receipts, edits are tracked, and a stale, failed or masked "all tests pass" is turned into a rerun instead of an ending.
dsh-wallpaper
chinarxq/dsh-wallpaper
Wallpaper skin for the DSH Web UI: image background with opacity, mask and blur controls.
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.