Plugins

Let the Blue Big Fat Fish browse Bilibili for you: it comments on your behalf, DMs you interesting finds, writes a diary, and keeps a memory for long-term summaries. It filters videos to your taste, creates Markdown digests, and can watch comments and reply for you. In Circle Mode, it replies only inside threads explicitly invited by the video uploader in a top-level comment — ordinary viewer requests do not count, and it never starts a new top-level comment. Commenting, DMs, auto-run, and comment watch are off by default. Needs Node 22, DSH 0.1.7-rc.1+, a signed-in Bilibili account, a model route and Python.

Reusable Agent Team presets for DeepSeek Harness: a Settings page stores named Teams — a captain with its own prompt and tool allow-list, plus members carrying their own provider/model route, reasoning effort, prompt and tools — and a composer control applies one Team to the current Session, installing the captain prompt and registering the spawn_team_member, get_team_preset and update_team_preset tools.

Standalone wallpaper background for the DSH Web UI and desktop app: built-in wallpapers, local image/video uploads stored in IndexedDB, and a read-only local Wallpaper Engine library served by the bundled wallpaper-engine-api service in the repository; renders images, video and scene workshop previews, with light-fog/dark-overlay, blur, brightness and safe-zoom controls plus a read-only follow-desktop-wallpaper mode.

将本机已登录的 Qoder（国内版 Qoder CN / 国际版 Qoder）模型接入 DeepSeek Harness —— bring locally signed-in Qoder models into DeepSeek Harness with zero configuration.

Orchestration-mode preset generator for DSH: a lead agent delegates to four built-in role helpers that carry only their own tools, and the web panel adds or removes custom helpers with their own tools, skills, and models, per workspace — zero changes to the harness source.

Remote access for the DSH Web GUI: LAN QR access, cloudflared quick tunnels, and frp reverse proxying via your own server, with a built-in auth portal.

Laya typed decisions (`noul` yes/no, `choice`, `score`) as a first-class Cordis service (`ctx.laya`) plus the model-visible `laya_ask` and `laya_plan` tools. The plugin itself installs and downloads nothing: it is a client of a `laya-mcp serve` sidecar you start yourself, and it reports that the sidecar is not running instead of failing obscurely at the first tool call.

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.

Model-autonomous context compaction for DeepSeek Harness: a mind_map tool surveys the agent context surface (stable seq ids, roles, token weights, clear-boundary markers, the clear-mind playbook), a clear_mind tool replaces a chosen conversation-history span with a model-written checkpoint through the native compaction transaction, call and result pairs self-collapse into one-line tombstones, and a proactive reminder nudges the model when the context grows long or a turn runs too many steps.

Let text-only models see images: when a picture arrives with a placeholder like \[image omitted because this model accepts text only], the model calls the vision_describe tool and the plugin forwards the image reference plus the question to a multimodal model, retrying with a fallback model on failure. Configured in the settings page and persisted via the official settings API.

Takes over the edit/write rows in the chat stream with the official native row shell and collapse behavior; expanding shows per-file code cards with file-type icon, add/delete stats, per-block collapse, highlight/plus-minus toggle, syntax-highlighted unified diff with real file line numbers, and a configurable context margin. Also ships a patch tool (beta, off by default) that applies multi-spot multi-file patches atomically via the *** Begin Patch envelope.

Same-origin in mechanism with the two loudest lines in context compression. The code-skeleton gate follows the skeletonization approach of Headroom (Apache-2.0), whose published headline is 20% fewer tokens for coding agents and 60–95% fewer tokens for JSON, same answers. The estimator channel follows TokenPilot (arXiv:2606.17016), which reports up to 60% lower cost for long-session agents. Both figures are theirs, quoted as-is; this plugin ships no benchmark of its own and claims no reduction of its own. What it adds for DeepSeek Harness: choose a compression profile, set the Auto Compact trigger level and toggle code-skeleton compression from one settings section, with exact DeepSeek V4 tokenizer measurement, same-revision count verification, and fail-open behaviour that keeps the original tool results on unsupported models. Within a turn it supports skeletonized information compression with a low original-text retrieval rate; at the tail of a turn it supports a turn-tail intent summary implemented purely by content substitution; context content stays safe, and the whole process runs inside the official JSONL compression.

Routes model requests, web fetches/searches and spawned tools through the Windows system proxy or a custom address — idempotent kernel (reads never rebuild the dispatcher), header one-click toggle, per-URL route diagnosis and NTLM-aware 407 hints.

Replaces the stats line under the message input in the DSH desktop app with a configurable status bar of ten optional fields, with live cost estimates at DeepSeek's official peak and off-peak rates, custom per-model prices, a CNY or USD switch and account balance.

Quote selected conversation text into the Web composer as native reference chips for focused follow-up turns.

Connects DeepSeek Harness to Busabase: search workspace knowledge, work with structured records and apps, submit writes as reviewable ChangeRequests, and inspect results in live cards.

Pins every request to the native max-effort tier and adds a per-model/per-mode reasoning-depth panel with a 0-20 capability slider, isolated Velocity/Apex profiles, and offline reasoning toolboxes. Distributed as an obfuscated build; commercial use allowed, reverse engineering closed.

Token-efficient persona runtime for DeepSeek Harness with local catgirl rendering and progressive tool-schema disclosure.

Provides an OAuth account page for DSH Web, currently supporting OpenAI Codex, GitHub Copilot, Anthropic, Kimi For Coding, OpenRouter, and xAI.

Codex-style review of session file changes: per-turn cards (optionally taking over the shell edited cards) whose totals are the turn net diff, a review tab with syntax-highlighted diffs that never wrap, real line numbers, open-at-line and a draggable split, a hover preview of the whole file with three lines of context around each change, one-click per-turn revert, and the on-disk record size with a clear button in the review toolbar. Records are sharded per session, with content-addressed snapshot blobs that dedupe and are garbage collected. Light/dark presets follow the DSH theme, typography follows DSH and installed skins (or any font on the machine, with a px stepper), and a settings switch injects a think-in-Simplified-Chinese rule into the global AGENTS.md.

MySQL connector for DeepSeek Harness: configure multiple connections in the settings page with per-connection table allowlist and write permission, switch the active connection from the composer, and expose mysql_query / mysql_tables / mysql_execute tools to every agent preset.

Inline +N −M badges on edit/write tool rows and a collapsible per-turn file-change summary card with aligned diffs, review and undo — covers native edit/write, str_replace_editor and Code Dispatch sub-calls, no git required.

After each completed agent turn, a bounded auxiliary LLM call writes one suggested next prompt into the session log; the web composer renders it as ghost placeholder text, adopted via Tab.

Rotates configured LLM API keys after quota, rate-limit, or authentication failures.