Plugins

Stores the agent's long-term memory as plain Markdown in an Obsidian vault — every memory is a .md file with YAML frontmatter that Obsidian indexes natively, so you can read, edit, link and back up it with tools you already use. Nothing is locked in a database. Ships a sidebar panel (tag filter, sort by recency or importance, edit in place), 7 agent tools, three-tier token-budgeted recall, BM25 ranking (no embeddings, no API key, works offline, CJK bigram tokenization), importance decay with archiving (moved to .archive/, never deleted), and optional GitHub Contents API sync. A note containing a secret is refused rather than redacted, because a redacted memory reads as complete while the important part is gone. Zero runtime dependencies; everything comes from the DSH host.

Turns "restarting the host" into one interrupt that many sessions can share. Several sessions queue onto the same restart (8 by default, configurable, joinable at any time including mid-wait); the host is closed once and relaunched once; then each waiting session is woken separately with its own reason and starts running its own work without waiting for the others. Each session decides independently whether to resume, one failed delivery never affects the rest, and a session can drop only itself without cancelling everyone else's restart. Windows desktop only.

Idle-time context truncation for a single DeepSeek Harness session. After a configurable number of idle hours the older conversation is dropped without a summary, so one long-lived session cannot leak one task into the next; it ships a per-session toggle, a tab next to the chat, and a /truncate command that IM bridges can drive.

Cross-session reference cards: drag a Q&A card out of any session into another one and it stays there; when you ask a question in the target session, the dragged cards' question, the answer it got at the time and its attachments are injected into the prompt as earlier conversation context, so the model can quote them directly.

Injects a Chinese-thinking reminder around the system prompt and the per-step runtime snapshot, so the agent reasons in Simplified Chinese instead of drifting back to English as the transcript grows.

Compress every idle session's context at a fixed time each morning on DeepSeek Harness: a daily run (08:30 by default), a catch-up run on the next launch after a missed slot, and a /compact-all command for on-demand compaction, with per-session outcomes reported.

Persona state and slow evolution for DSH: five state dimensions (valence, arousal, energy, intimacy, focus) move after each turn and decay toward baseline over time; four scene types (work, tech, support, chat) modulate reply length and warmth without changing the state itself; the Soul panel keeps a persona card plus weighted evolution entries and injects them into the system prompt persona-first under a 4000-character budget. Ships a browser panel with a persona card editor and a live state view (the five metrics plus the last six events), appends a JSONL trace to disk, and seeds a usable default persona on first install.

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.

Turn a logged-in DeepSeek web account into an OpenAI and Anthropic compatible endpoint, with tool calling, image input, an account pool and a local admin console. First use needs a one-time `node scripts/fetch-assets.mjs` in the installed plugin directory to download DeepSeek's PoW runtime from the vendor CDN; without it the proxy starts but every upstream request fails.

Post-build patch for the maid-atelier skin, mounted as a bundle: the window backdrop walks six time-of-day bands on the local clock with continuously interpolated sky and interior light, and the figure moves on its own (breath, weight shift, float, blinks, optional hem and hair displacement); applied to an already-installed skin and reverted from client.js.orig when the plugin is removed.

Floating Token Burn Dock for DSH that aggregates provider usage across every session: the collapsed capsule shows the live tokens/sec rate with an inline 60-second sparkline and today's total, while the expanded panel adds a large live-rate hero, per-model and per-hour breakdowns on a real left axis, and a 30-day commit-style heatmap. The sequence-frame flame burns only while tokens are actually flowing and goes out at rate 0; day and hour buckets follow the browser's UTC offset instead of the host process timezone, and the 31-day record store is written atomically after history load so a restart no longer wipes the numbers.

Appends the current turn's estimated token cost in CNY to the assistant action row, right after the usage/time indicator. Prices come from a built-in DeepSeek peak/off-peak CNY table (Beijing time, Monday-Friday 9:00-12:00 and 14:00-18:00) and can be overridden from the profile's cordis.patch.yml; a model without a configured price shows a dash instead of a made-up amount. Follows the shipped Performance & usage - Compact setting and hides together with the usage pill. Token cost only, per turn only.

Shows how much of your Claude and Codex plan limits you have used: a line above the prompt in dst with 5-hour and weekly windows, percent used and time to reset, and a ring meter beside the context meter in dsh web that opens a panel listing every window.

Windows resource monitor for the DeepSeek Harness (dsh) sidebar: samples CPU, memory and clock speed once a minute, keeps a 30-day history and draws separate 24h / 7d charts for CPU and memory (with 85% / 90% alert lines) plus a frequency chart that marks load-time throttling. Reports sustained CPU >= 85% (10 min), memory >= 90% (10 min) and frequency <= 70% of maximum clock under load (5 min).

Live step pill in the DSH session header: folds the durable session event stream into the agent's current action (tool name plus its argument summary), the step index inside the running turn, the elapsed turn time and the todo progress, polls it every 1.2 seconds, and renders blue pulse while working, green for a finished turn and red for a failed tool call; hides itself 9 seconds after a turn ends. On Windows it also opens an always-on-top desktop HUD (set DSH_AGENT_STATUS_HUD=0 to disable).

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.

Fairy III (Zenless Zone Zero) H.D.D skin for DeepSeek Harness: boot video, an eye-core dock that can be dragged vertically so it never covers text, and time-of-day greeting voices written by a model every day (5 lines, each <= 20 characters) and synthesized locally with GPT-SoVITS, with a fallback chain of today -> latest -> bundled audio so there is always something to hear. Replies can be spoken: a trailing marker line takes priority over length heuristics, and long replies can be compressed into one sentence first. The reader hashes the utterance it just played (MD5) so each reply is voiced exactly once, and switching sessions stays silent. Personal fork of dsh-fairy-zzz with its own package, route and storage namespace, plus a rate-limited summarize endpoint, POST-only engine control and bounded TTS and greeting caches. Two diagnostic logs (client revision, decision trace) are written next to the plugin and stay on the local machine. A personal fork of AssassinVIX/fairy-zzz-plugin (upstream package dsh-fairy-zzz): install this fork or the upstream, not both - both mount the full-screen skin on the same DSH surfaces. Dynamic greetings and spoken summaries need a DeepSeek-compatible API key in ~/.dsh/fairy-tts.json (or DEEPSEEK_API_KEY); without it only the bundled greeting audio plays.

Renders mermaid code blocks in dsh web chat as diagrams. Each block gets a toggle in its header to switch between the diagram and the source; blocks still streaming stay as code until the message settles. Mermaid loads lazily the first time a diagram is shown. The terminal UI is unaffected.

A split-flap salary board for the DSH Web UI. Enter monthly salary, hours per day, working days per month and a start time, and the board settles today's earnings every 2 seconds on eight flip cells (currency symbol, four integer digits, decimal point, two decimals). Shows 0.00 before the shift, caps at the daily wage after it, and handles shifts that cross midnight. Six color schemes, drag anywhere (the position is kept on this machine), and a close button that collapses the board into a pill in the session header. Pure client-side bundle: no network calls and no dependencies.

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.

Skill bundle for tiered multi-agent science campaigns in math, physics, astronomy, engineering, and biomedical evidence: a shared blackboard, independent verifier and red team, wolf-pack and flanking tactics, token and time budgets, known-answer evals, a four-agent pendulum pilot, and a six-agent fourth-order pilot that caught and corrected a benchmark answer-key error. Does not claim open-problem solutions.

Keyless image generation with two tools: generate_image (no API key, one call at a time, bounded rate-limit retries) and list_image_models.

Starts the published local BeatDesign 0.2.5 workspace and connects 30 MCP tools for projects, Canvas, Assets and Timeline, with a shared-project Skill and remote-generation approval guidance. Requires Node.js 22.12+ and a free port 3020; frame extraction and timeline rendering also need ffmpeg/ffprobe on PATH (browser export does not).

Master Duel utility panel for DSH: card lookup with banlist filters and AI-assisted effect search, related cards, rarity badges, favorites, a deck builder with drag-and-drop and a collapsible favorites panel, engine-tested opening hands and hand_odds opening probabilities, deck-type rankings, and deck checking. Card images and text are fetched from third-party sources at runtime; the plugin bundles no card data and requires the separate master-duel-ai-advisor MCP server.