플러그인

A draggable floating card for the DeepSeek Harness Web UI that combines a Pomodoro timer, an off-work countdown, and focus statistics. Each launch opens on a fresh focus phase; phase completion triggers a chime, card shake, and particle burst. Includes today's pomodoros, focus minutes, streak and a 7-day chart, rhythm reminders for wrapping up and eye rest, optional system notifications, tab-title timer, auto-start, +5 minutes, and a custom long-break interval.

Wall-clock session and turn timing for the DSH Web GUI: a timer chip in the session header shows the session total, every finished turn shows its own duration above the action row, and a sidebar panel totals today, the last 7 days, the last 30 days or all sessions. The numbers are folded from turn/start and turn/end in the session log, so paging, compaction and restarts do not change them.

A countdown pill under the message composer showing how long is left before you clock off, with a workday progress bar; click it to set your own start and end times (saved in this browser), and after the end time it turns into an overtime timer.

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. 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.

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.

호스트 모델과 대화하고 클릭 시 애니메이션을 재생하며 타이머에 맞춰 속삭이고 작업 진행을 말해 주며 계정 잔액을 표시하는 투명한 항상 위 데스크톱 펫입니다. DSH Desktop의 번들 Electron 런타임(~/.dsh/electron)에서 작동하며 런타임이 없으면 펫이 나타나지 않습니다. Windows 전용입니다.

DeepSeek Harness Web UI용 뽀모도로 타이머입니다. 집중/짧은 휴식/긴 휴식 주기가 있는 사이드바 페이지와 composer 옆 카운트다운 pill을 제공합니다.

DeepSeek Harness용 Zen mode입니다. 작업 중 DSH를 숨기는 전체 화면 미니멀 타이머를 보여주며 경과/남은 시간, 해당 라운드에 완료된 세션, 종료 버튼만 표시합니다.

작업 중 잠시 멈춥니다. 현재 step을 마친 뒤 대기하며, 재개해도 새 request를 보내거나 토큰을 쓰지 않습니다. 일시 정지 중 timer는 멈추고 현재 session에만 적용됩니다.

채팅 composer에서 보낼 시각을 지정하면 해당 시각에 Host가 메시지를 전달합니다. 기존 대화에서 예약하면 작업이 그 대화에 들어갑니다. 새 세션 화면에서 예약한 각 작업에는 세션이 미리 생성됩니다.

DSH Web GUI용으로 드래그할 수 있는 플로팅 시계입니다. 아날로그/디지털 표시, 집중 타이머, 테마 4개를 제공하며 배치를 저장합니다.

composer 옆 버튼으로 현재 초안을 더 실행 가능한 prompt로 바로 다시 씁니다. 가벼운 다듬기부터 적극적 확장까지 기본 preset 4개와 사용자 preset을 지원합니다. composer를 떠나지 않고 버튼에서 preset/prompt model/추론 강도를 바꿉니다. 초안에 내용이 있을 때만 보이고 강화 중 타이머와 실패 시 재시도를 표시합니다.

DSH 턴 타이머입니다. 현재 턴 실행 시간을 알리고 `/set-time` 일정에 따라 마무리 prompt를 주입해 에이전트가 계속 실행하지 않고 끝내도록 합니다. composer 위에 실행 시간 카드도 표시합니다.

4분면 작업이 있는 Pomodoro 타이머입니다. 집중 라운드를 작업에 연결하고 EXP와 업적으로 데스크톱 펫을 키우며 에이전트가 라운드를 시작하거나 작업을 제안할 수 있습니다.

Move45clock은 DSH 건강 시계다. 45분 타이머(5–120분), 12가지 앉거나 서서 하는 12–15분 휴식 계획, 시각적 계획 편집기, 200%까지 확대 가능한 알림 카드, 21:30/22:00/22:30 야간 알림, 선택형 동작명 음성 코칭을 제공한다. 창 깨우기는 Windows 전용이다.

유휴 상태 또는 타이머에 Akasha sleep 증류를 예약하고 새 세션에 기상 노트, 각 턴에 메모리 펄스를 주입한다. 유휴 후 재개는 기본 꺼짐. 연동: Hindsight(repo memory), OpenViking(user memory), 내장 session search(repo README 참조). 설정: repo clone, core/ 별도 복사, cordis.patch.yml의 <AKASHA_DIR>를 절대 경로로 바꾸거나 AKASHA_DIR 설정. 표식이 남으면 명시적으로 실패한다.

DSH용 표준 25/5/15 뽀모도로 타이머다. 각 실행을 현재 세션 작업 또는 직접 입력한 프로젝트명에 연결하고 통계 패널, 365일 히트맵, 드래그 가능한 플로팅 타이머와 모서리 표시를 제공한다. 로컬 JSON 저장 및 CSV 내보내기를 지원한다.

부담 없는 컴패니언 패널을 추가합니다. 포모도로 타이머, todo 목록, 물 마시기·일어서기 알림과 설치된 데스크톱 펫을 통한 수업 설명 및 숙제 지도를 제공합니다.

드래그 가능한 플로팅 창에서 지뢰찾기를 즐깁니다. 첫 클릭 안전 배치, 깃발, 숫자 주변 동시 열기, 타이머를 제공합니다.

DeepSeek용 요금 인식 작업 실행기입니다. 비피크 요금 시간에는 프롬프트 묶음을 deepseek-flash로 바로 보내고, 피크 시간에 가까우면 자체 타이머로 대기열에 넣었다가 다음 비피크 시작 시 실행합니다. peak_run, peak_schedule, peak_cancel을 제공하며 피크 시간대와 선행 시간을 설정할 수 있습니다.

확정적 답을 내는 가장 저렴한 검사 방법을 선택하는 검증 skill입니다. 정적 검사부터 단일 테스트 스위트까지 L0~L3 비용 단계, 고정 sleep 대신 타임아웃이 있는 조건 폴링, 실제 실패할 수 있는 assertion, 기억이 아닌 실측 시간을 규정합니다. skill이 쓰는 무의존성 Node 도구 2개도 포함합니다: `wait-until.mjs`(명령, 파일, 포트가 준비될 때까지 폴링)와 `measure-runtime.mjs`(모든 명령의 실시간 측정 및 라이브러리용 구간 타이머).

DeepSeek Harness용 앉기/물 마시기 알림입니다. Web GUI 구석에 고양이를 표시합니다. 30~90분 집중 주기가 끝나면 중앙 오버레이와 3회 알림음으로 일어나 걷고 물을 마시도록 알립니다. ‘later’ 버튼은 오래 앉기와 탈수에 대한 출처 있는 실제 사례를 각각 하나씩 보여줍니다(발행처, 날짜, 원문 링크가 있는 17개 항목 내장). 중지 시 작업 시간, 완료 주기, 성공한 중단, 미루기, 권장 음료, 좋음/보통/나쁨 등급을 포함한 세션 보고서를 냅니다. 고양이는 끌어 옮길 수 있고 위치를 기억하며 상태별 애니메이션을 반복합니다. 타이머와 기록은 호스트 측에 있어 페이지를 새로고침해도 주기가 유지됩니다. DSH 페이지가 열려 있을 때만 알립니다. 주기 길이, 알림음 크기, 고양이 불투명도, 등급 기준을 설정할 수 있습니다.