Plugins

Failure memory for DeepSeek Harness. Captures tool failures as they happen, classifies them with 16 built-in rules into lessons pairing a symptom with a fix, adds session/daily/weekly summaries, and injects a short briefing into later sessions through systemPrompt context. Ships 5 pitfall_ tools and a web settings page.

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.

Local Ollama worker tools for DeepSeek Harness: model inventory, a single unverified ask, and a batch runner that chunks long text, checks every item deterministically, returns a small verbatim sample for review and can optionally write the labelled result inside the session workspace.

Named model routes for subagent delegations, configured from a settings page: each alias binds a subagent provider, model and optional reasoning effort, token cap, agent preset, tags, plus a "when to use" line the model reads to pick it, so a delegation names only the alias. Adds five tools (delegate, list routes, edit config, wait, release), shows the route a child actually ran on in its session header, lets each UI element be switched off, and keeps all configuration in one local JSON file instead of YAML.

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.

WYSIWYG Markdown notes editor in the DSH right sidebar, with tabbed documents, table editing, chat integration, assistant read/write tools, and local file export.

WeChat Reading (WeRead) right-sidebar bookshelf and chapter reader: QR or Cookie login, store search, resume reading, and weread_shelf / weread_search / weread_book / weread_chapter agent tools.

A composer control: shows the current model's context window, the tokens in use and a rough system / tools / messages breakdown, switches the model's reasoning effort in one click (writing reasoningEfforts into the model config when the provider declares none), and edits the model's contextWindow from the same panel so custom-provider models stop silently inheriting the adapter default. Makes no model calls; reads only the official session projections, model directory and config-forms services.

Optional source-preserving chat rendering with selectable themes, offline Uiverse component tools and a report-local evidence workspace. Preview for DSH 0.2.1-rc.2.

把你和 AI 的全部对话，渲染成一片星海，可视化的预览

Remote Linux server manager in the Harness Web UI: browse and edit files, inspect host status, start/stop/restart Docker containers, inspect images, configure Docker registry mirrors, manage firewalld/ufw rules, switch apt sources, and detect developer tools or install/upgrade them with apt.

Adds a Git chip beside the composer that opens a centered workbench (changes, history, files, tags, stash) plus a Git section in settings. Supports multi-account OAuth device flow and PAT with per-repository permissions (full, read, none), gates writes behind a global switch and reads behind an allowed-directory list, and refuses a push whose target remote belongs to no writable account. Diffs preview in place with line numbers, four layouts, word-level emphasis, syntax highlighting and Ctrl/⌘+F find; Markdown and images render inline, commit messages are searchable, and commit messages can be generated from the staged diff through the host LLM service. Ships 8 Agent tools, 51 local HTTP endpoints and 8 languages, with tokens encrypted at rest by AES-256-GCM.

Delivery workflow preset for DeepSeek Harness: an agent protocol plus seven tools that keep user requests verbatim, record feedback decisions, plan tasks with acceptance criteria, log real verification commands and exit codes, and write completion summaries as per-project JSONL files.

Fully local push-to-talk voice input for macOS: hold a global hotkey, speak, release — transcription is pasted at your cursor. Local-first STT (Qwen3-ASR 0.6B 4bit on MLX/Metal by default, SenseVoice and mlx-whisper fallbacks), zero external APIs, launchd self-healing service and an 8-point diagnose script. The bundled skill lets the agent install, configure (Karabiner hotkey wiring, launchd, engine switching) and troubleshoot the whole toolchain for you.

Prepares files before read, read_image and the vision tools open them: downscales images over 2048 px or 8 MB (one retry at 1280 px within the same turn), converts BMP/TIFF to PNG, relabels mislabeled uploads by sniffing their bytes, and turns docx/xlsx/pptx, text-layer PDF, csv/tsv and html into Markdown with no external programs.

Two PDF tools for the DSH sidebar in one install — jump to any page and browse the whole document with locally rendered pages and thumbnails, or open a PDF with DSH's native preview. Ships a /pdf command too.

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).

Looks up CCF conference paper deadlines with the ccf_deadlines, ccf_venue and ccf_update tools, the /ccf command and a desktop panel that exports the list as a calendar. Dates come from a live sync (CCFDDL as the main source, a second source cross-checked; a gap of 36 hours or more is flagged as a source conflict), so nothing is bundled and the next 30 days stay in the system prompt.

Read-first VPS operations tools: systemd status and actions, disk usage, Docker status, log tails, configured backups, and quiet-hours schedules.

Agent tools for a local Gopeed downloader: create, inspect, pause, resume and remove downloads, diagnose or launch the Gopeed process, and receive download-complete events through a DSH background job that wakes an idle agent. Process diagnostics and launching are Windows-only; the download tools work wherever Gopeed runs.

Deny-only tool blacklist per agent preset. Tick a tool off in Settings and it leaves that mode's tool schema and its calls are rejected; nothing is ever enabled or whitelisted, and tools nobody denied are left alone. Each row expands to that tool's own description, and rules live in one JSON file outside the profile.

Memory for DeepSeek Harness with five sections (memories, tree, profile, graph, inbox), a rule-based capture policy that decides what to keep and what to drop, and hybrid BM25 plus vector retrieval over parent and child chunks, backed by twelve agent tools that read and write the same local files.

Local-first research toolkit for DSH: registers kb_grep, kb_list, and kb_save model tools. kb_grep searches the markdown knowledge base with ripgrep when present and falls back to grep, while kb_list and kb_save go through the scoped ctx.fs filesystem service; the KB defaults to ~/dsh-kb. The repo also ships two skills (local-first-research and session-memory) that must be copied into a DSH skills directory yourself; they are not installed by the plugin bundle. The plugin is pure Node ESM with no npm runtime dependencies and ships a framework-free node test harness. Installable via plugin_manager install_bundle; no client UI is bundled.