Plugins

Floating Pomodoro dial pinned to the bottom-right of the Harness UI: custom focus, short-break and long-break lengths, a long break that can be switched off entirely, a round counter, and five Web Audio-synthesized alarm sounds with preview, volume and ring duration. No audio assets and no network requests; timer state survives a reload.

ADHD-friendly focus soundscape docked under the DSH composer: five procedurally synthesised noise colours (white, pink, brown, rain, fan), a live-waveform orb, a 15-90 minute fade-out timer and a liquid-glass popover. No audio assets and no network requests.

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.

A pomodoro timer docked below the composer in DeepSeek Harness, with focus/short/long cycles, a resumable countdown, chime, optional system notifications and bilingual settings.

A transparent always-on-top desktop pet of Firefly (流萤) that chats through the host model, plays animations on click, whispers on a timer, narrates task progress, and shows account balance. Runs in the DSH Desktop app using its bundled Electron runtime (~/.dsh/electron); the pet does not appear if that runtime is missing. Windows only.

Pomodoro timer for the DeepSeek Harness Web UI: a sidebar page with focus/short-break/long-break cycles plus a countdown pill beside the composer. · DSH Web 番茄钟：侧栏整页（专注/短休息/长休息）+ 输入框旁倒计时胶囊。

Zen mode for DeepSeek Harness: a full-screen minimal timer that hides DSH while tasks run, showing only the elapsed or remaining time, the sessions that finished during the round, and an exit button.

Take a breath mid-task: the current step finishes, then it holds. Resume sends no new request and spends no tokens; the timer freezes. Current session only.

Timed send for the chat composer: pick a moment and the host delivers the message then. Inside an existing conversation tasks join that conversation; from the new-session screen each task gets its own pre-created session.

A draggable floating clock for the DSH Web GUI with analog and digital faces, a focus timer, four themes, and persistent placement.

Button beside the composer that rewrites the current draft into a more executable prompt in place. Ships four built-in presets spanning light polish to aggressive expansion, accepts user-defined presets, and switches preset, prompt model and reasoning effort directly from the button without leaving the composer. The button appears only when the draft has content, shows a running timer while enhancing, and surfaces failures with a retry.

Turn timer for DSH: reports how long the current turn has been running, injects wrap-up prompts on a /set-time schedule so the agent starts finishing instead of running on, and shows run-time cards above the composer.

Pomodoro timer with four-quadrant tasks: bind focus rounds to a task, grow a desktop pet with EXP and achievements, and let the agent start rounds or propose tasks.

Move45clock: a health clock for DeepSeek Harness. A 45-minute work timer (5–120 minutes), 12 guided 12–15 minute seated/standing break plans, a visual plan editor, reminder cards scalable to 200%, night reminders at 21:30/22:00/22:30, and optional action-name voice coaching. Window wake is Windows-only.

Schedules Akasha sleep distillation on idle or timer, and injects a wake note for new sessions plus a per-turn memory pulse; the optional idle-resume behaviour is off by default. Companions: Hindsight (repo memory), OpenViking (user memory), and the built-in session search (see repo README). Setup: clone the repo, copy core/ aside, then replace <AKASHA_DIR> in cordis.patch.yml with its absolute path (or set the AKASHA_DIR env var); plugins fail loudly while the placeholder remains.

Classic 25/5/15 pomodoro timer for DSH that ties each run to the current session task or a manual project name, with a stats panel, a 365-day heatmap, a draggable floating timer and a corner pill, local JSON storage, and CSV export.

Adds a low-key companion panel with a pomodoro timer, a todo list, drink-water and stand-up reminders, plus lesson explanations and homework tutoring through the installed desktop pet.

Minesweeper in a draggable floating window with first-click-safe layout, flags, chord and timer.

A price-aware task runner for DeepSeek: during an off-peak pricing window a batch of prompts goes straight to deepseek-flash; near peak it queues the batch behind a self-held timer and runs it at the start of the next off-peak window. Exposes `peak_run`, `peak_schedule` and `peak_cancel`, with configurable peak windows and a lead time.

Ships a skill for verifying a change with the cheapest check that still yields a definite answer: an L0-L3 cost ladder from static checks to a single suite, condition polling with a timeout instead of fixed sleeps, assertions that can actually fail, and durations that are measured rather than remembered. The package also carries the two zero-dependency Node tools the skill calls - wait-until.mjs (poll a command, a file, or a port until it is ready) and measure-runtime.mjs (wall-clock timing for any command, plus a segmented timer for library use).

Sit-and-water reminder for DeepSeek Harness, with a cat that sits in the corner of the Web GUI. A focus cycle of 30-90 minutes ends in a centered overlay plus a triple chime that gets you up to walk and drink; the later button answers with one real sourced case about prolonged sitting and one about dehydration (built-in library of 17 entries, each with publisher, date and a link to the original); stopping produces a session report with worked time, completed rounds, successful interruptions, snoozes, suggested drinks and a good/mid/bad grade. The cat can be dragged anywhere with its position remembered and loops one animation per state. The timer and the ledger live in the host half, so a page refresh never loses a cycle; it only reminds while a DSH page is open. Cycle lengths, chime volume, cat opacity and grade thresholds are configurable on the plugin settings page.

Session-local scheduled tasks: inject a prompt into the current session after a delay, at a fixed time, or on a repeating interval, so the model wakes up and runs it on time. Manage them from a panel above the composer, the /schedule command, or seven model-facing tools. Timers freeze while DSH is closed and never auto-resume; pause keeps the remaining time, so resuming continues from where it stopped.

Explains which layer killed a failed turn: times every model request (first byte, longest silence, total) and combines that with the error to name the culprit (dsh idle watchdog vs Node undici timers vs SDK timer vs server crash, context overflow, gate queue, tool timeouts, failed compactions) and the exact setting to change; verdicts in the DSH terminal plus a /why command, subagents included.

Scheduled tasks and message push manager: cron scheduling with auto push to Feishu and multi-channel, dual-env (OpenClaw/DSH).