Plugins

A Bilibili companion that keeps its own schedule. A 60-second tick in the plugin host wakes a dedicated agent session at a random moment inside the configured time windows (opting out of a day is a configurable chance), then hands it a bundled CLI of 38 subcommands — feed, subtitles, danmaku, comments, ffmpeg frame grabs, and two diaries. Shares at most N items a day by commenting on them and @-mentioning the owner (default 3, enforced from a written quota log, and it stops on rate-limit or captcha signals). Also polls comment replies, direct messages and live-room danmaku; replies to anyone but the owner are queued for approval in Settings. Settings → 养鱼 writes ~/.dsh/bili-whale.config.json; cookies live separately in ~/.bili-ai/credentials.json and are never echoed back. The HTTP routes are loopback-only behind DSH's trust fence. Needs Node 18+ and ffmpeg on PATH for frame extraction.

QQ official bot channel (enhanced fork of tencent-connect/dsh-qqbot): full agent chat in groups and DMs, rich media, an auto-tagged sticker library with AI image replies, join approval, mute and broadcast, scheduled wakeups, remote approvals over QQ, interactive cards, multi-account instances, extensions, member profiles, and token-saving value-scored wakeups.

Session-level superagent: watch one GitHub repository and wake this session when an issue or pull request changes; requires a logged-in gh CLI, and the credential never reaches the model.

QQ official bot channel for DeepSeek Harness (feature-rich fork of tencent-connect/dsh-qqbot): group and private chat with the full agent, sticker library with auto-tagging and AI meme replies, group management (join approval and mute), botplay interactive card events with paginated catalogs, user extension registry (custom slash commands and QQ tools), scheduled wakeups, remote approvals over QQ, and a floating dock web UI.

Full-duplex voice mode for the DeepSeek Harness Web UI: toggle (2s-pause auto-send) or hold-to-talk dictation into an editable draft with zipformer2 streaming ASR, optional wake word; sentence-by-sentence Edge TTS read-aloud with live captions, and speaking interrupts playback and the running turn (true barge-in). On-device ASR, no API key.

An agent that wakes up on its own schedule: by default every 60 minutes, ±20% (48–72). It browses the interests you set, drops what it finds into a local pool of notes, and keeps a profile of what you like. When it has enough worth sharing it hands the session you linked a short package of notes — whether to speak, and which note to use, is that session's agent's call in the moment. Quiet hours, busy windows, a daily limit and a cooldown are decided in code; when the answer is no, the decision and expression phases after the gate call no model at all — browsing runs before the gate and is counted separately. Windows, Linux and macOS (Linux and macOS are minimal support).

Condition-driven wakeup: durable file/command/http/process/webhook watches that wake the agent.

Adds an animated ASCII whale ocean behind empty chats, with neutral frosted-glass input, visual color selection, light/dark support, and configurable whales, fish, grid and wakes.

Host-level proactive wake-ups for DeepSeek Harness: the model schedules its own alarms (once, recurring interval, or cron, with optional random jitter), cold sessions are woken on time by session resume, model-initiated follow-ups are gated by quiet hours and a daily delivery budget, silent wake turns end with proactive_reclaim and fold out of model context as tombstones, declared schedule files glob-sync JSON schedules into host alarms idempotently, min_idle_seconds defers wakes until the destination session quiets down, and a web panel (settings section plus per-session tab, SSE live refresh) manages alarms and wake history — the heartbeat engine for AI companions and roleplay characters with their own daily routines and world evolution.

A character who lives on her own phone: the plugin installs and runs her Android phone, and any app notification can wake her; she installs apps herself and browses, likes and comments in apps such as Xiaoheihe or Bilibili; she has her own QQ (NapCat / SnowLuma, multi-part replies, stickers); memory with people cards, a daily diary at 23:30 and automatic recall before each wake (local embeddings by default, SiliconFlow / DashScope optional); a life between conversations (after HDS-Interlude); and an optional microVM computer with approval gates. Sidebar tabs show her phone, her computer and a dot panel; personal settings live in ~/.dsh/dsh-dot.json, and each of the six components can be turned off. Only the QQ number set with /qq owner counts as you; other people's messages are treated as untrusted notifications.

Old conversation is folded into short summaries by a side model before the window fills (on-disk ledger survives restarts and re-anchors after host compaction), so long sessions stop getting more expensive every turn — one real full-day session: 6 folds replaced 209 old messages and the live context stayed in the tens of thousands of tokens. The session model picks the model per subtask across six pools (economy/mechanical/main/hard/vision/review); or pick Auto and a small intent classifier routes each message itself, switching only past a confidence threshold to keep prompt caches warm. A three-line watermark ends in an automatic four-step handover (fork backup, handover doc, compact, wake) without losing in-flight subagents; /ctx-pause, /ctx-resume and /ctx-handover keep it manual when wanted. Optional Agent Teams mode adds teammates and a shared task board; /vision gives text-only models eyes via the vision pool; tiered verification sends bigger diffs to a tester plus an independent reviewer. Settings UI, READMEs and language preferences cover the same 11 languages. Bundles three skills: read-only MySQL/Redis verification, commit-message generation, and a requirements/design document convention.

Connects one existing DSH session to QQ through OneBot/NapCat without starting a new session, and uses a fail-closed whitelist plus @-mention, nickname and configurable-probability wake rules to decide which messages reach the model.

Wakes the main agent after any DSH restart so interrupted work resumes automatically — adds a restart tool and an optional canary that validates the boot and aborts with an alert before a broken restart.

Local cross-session messaging for DSH agents, with discovery, cold-session wake-up delivery, and auditable message history.

Model failover and turn keep-alive for DeepSeek Harness: a request that fails with a model error before any visible output falls through the models of the configured provider groups (then the next provider) until one streams; transient failures retry with backoff under a liveness watchdog, a dead turn self-resumes with templated 「继续」 wake-ups, and an optional full-auto mode auto-answers permission, question and plan approvals with every decision appended to an AUTO-MODE.md audit log.

Runtime extension suite for DeepSeek Harness: cross-session mailbox with wake cold-start, agent teams (captain + members + dependency task board + team_wait), subagent spawn policy with escalation approval, task-aware routing preset, plugin market, skill manager and runtime injector.

IM channel bridge: WeChat (ilinkai) / QQ / Feishu with proactive push — wake the channel bot from scheduled tasks and deliver AI replies to your phone.

Plugin that turns a XiaoAI (Xiaomi) smart speaker into a voice assistant: hosts the local bridge process, listens for the wake word, runs speech recognition (ASR / STT) on what you say, keeps the conversation in the host session, and speaks the answer back through the speaker with text-to-speech (TTS) playback.

Voice assistant for dsh web: say the wake phrase (e.g. "小鲸") to activate hands-free dictation — what you say is transcribed and typed into the chat box automatically. Supports spoken edit commands (send, clear, new line, stop reading) and reads assistant replies aloud in Chinese. Speech recognition runs locally in-browser via sherpa-onnx WASM, so it works offline without an API key.

Persistent background watchers that wake the agent on new messages (file inbox or command-output delta) — the harness analog of Claude Code's Monitor tool.

Group chat rooms for DSH sessions with explicit @mentions, cold-session wake-up delivery, and optional cross-host transport through dsh-weave.

Background stream listeners that wake the DeepSeek Harness agent — put a watch on a command or file and receive new lines as in-session notices. Ships a daemon host so a watcher can run unattended for weeks.

Wenyuan Pavilion: an AI Team workbench for DSH with owned tasks, independent acceptance, an append-only ledger and concurrent projects. Install it into its own profile (dsh plugin --profile ai-team add @suifei/dsh-ai-team), not web or desktop: its patch sets that profile's sandbox to danger-full-access and approvals to never, and the member agents run with full access. The project's PMO acts automatically by default (every 60 s it resolves pending decisions and wakes owners to dispatch the next tasks); turn off PMO auto to send decisions to your inbox.

Layered context compression for DeepSeek Harness: folds old turns into model-written L1–L3 summary blocks, maps what each layer holds, and wakes the model when the context crosses a size line or keeps growing. By default it requires DSH's experimental Session Search plugin (session_event_read / session_event_search); without it the tools stay disabled unless requireSessionQuery is turned off.