Plugins

Personal QQ channel for dsh: managed NapCat with QR login, private/group AI replies with batching and image understanding, local archive with polling backfill, in-group @ and self-recall, QQ Zone guestbook tools, and a sidebar panel. By default any QQ user who DMs the account (or @s it in a group) gets an AI reply; set access.privateMode / groupMode to allowlist to restrict. Owner-only commands stay owner-only.

Sidebar category groups for DSH workspaces: the flat project list becomes collapsible named sections, workspaces re-file by dragging a row onto a group or picking one from a dropdown, one toolbar button collapses every group at once, an appearance panel sets font size, text color and background tint for each of the three levels, and Ctrl+Tab cycles recently opened sessions.

The durable-agent engine for DeepSeek Harness: a host-plane cordis service mounting @earendil-works/pi-durable (conversations, exactly-once submissions, fork-at-entry, tasks) over a SQLite storage backend with an OpenAI-compatible relay provider.

Multi-channel IM bridge for DeepSeek Harness: each QQ (SnowLuma / OneBot v11, official QQ Bot) or Discord group and private chat becomes its own agent session, bound to a role card whose domain carries knowledge bases with draft approval, shared memory, a read-only GitHub reader and group trigger rules; two built-in personas (customer support, group chat), unified push with broadcast, and human hand-off.

Artifacts-like documents in the sidebar: the agent writes a self-contained HTML document, you comment on any part, send the comments back for the next version, or edit text and formatting yourself. Every change is a new version; local only. A local-only fork of tdoc (https://github.com/tornado-doc/tdoc), rewritten as a DSH plugin.

Layered long-term memory for DeepSeek Harness: each turn injects a persona frame plus a relevance-ranked memory index while full transcripts stay on disk and are expanded on demand, so the vault never floods the context window. Facts are slot-keyed, so a changed mind retires the old value instead of recalling two contradictory answers. The persona frame is derived from memory by the model itself, behind three anti-pollution gates (rejecting development chatter, blocking echoes of its own injection, and merging restatements by shared content words). Periodic consolidation merges near-duplicates, rolls long topics into digests, and archives low-value old entries by tagging only — raw text is never deleted. The append-only store is compacted in place (measured 4.7 MB to 355 KB). Zero third-party dependencies, no build step. Note: per-turn injection inevitably lowers prompt-cache hit rates, so this plugin is not recommended for work scenarios.

Complete mobile PWA for DeepSeek Harness (DSH): secure remote access gateway + install-to-homescreen + offline + touch gestures + agent-done push. deepseek-harness-plugin

Runtime plugin firewall for sessions started through its launcher (the DSH Desktop app cannot be injected and is not covered): filesystem, network and subprocess calls plugins make are attributed to the plugin that made them and judged against what that plugin declares, so a plugin doing its declared job stays silent and leaving it is recorded. Attribution is best-effort, not a guarantee - uncovered calls fall back to host, never to another plugin. Also adds a pre-install health card (capability surface, red-line combinations, PASS/WARN/BLOCK), boundary judging, opt-in pre-execution blocking with a credential-exfil rule, update diff with a trial install in an isolated child process, SARIF/CSV audit export, and multi-source aggregation. Local-only storage, no telemetry; the optional webhook is off by default.

Browses Bilibili on your behalf and actually takes part: search videos, like, coin, send danmaku, post comments, reply to other people, and keep replying when someone answers back. Reads the video's Bilibili AI subtitle (or transcribes the audio locally when there is none) so its comments talk about what the video really says, and it reads that same content before choosing whom to reply to. Instead of copying other people's sentences, it distills rules about how to sound like a person and gets less AI-ish the more it runs. Skips videos it already browsed, de-duplicates replies per thread, and after too many rounds with the same person in one comment section it reviews whether the exchange has any substance -- if it looks like two bots complimenting each other, it says one last line and stops replying to that person there. Checks whether each comment is genuinely public or only visible to your own account, can scan any time window (last hour / day / 3 days / all) and lists the ones others cannot see, warning you on the overview when replies got folded. Six-tab settings panel where every change takes effect immediately with no restart, a manual run that queues behind the background listener instead of being refused, and an environment self-check that tells you exactly what to install. Requires Python 3 for the Bilibili actions; the built-in check lists any missing packages.

Install it by asking DSH in one sentence, then it reads your local Codex CLI/Desktop chat history from ~/.codex/sessions and makes it usable inside DeepSeek Harness: a panel under the composer lists sessions with keyword and date filters and one-click import, the codex_history tool lets an agent list, read, search and summarize past conversations, a session whose working directory matches is injected automatically when a new agent starts, new Codex messages arrive as incremental updates in the running conversation, and a shared folder can mirror both Codex and DSH sessions across machines. Read-only — nothing is written back to Codex.

Control plane for the harness's own Agent Teams: members are ordinary independent sessions created through the same seam the GUI uses, each with its own model, so they open and continue like any other conversation while a sidecar roster and task board coordinate them. Task entries carry a definition of done, a member's completion is submitted and waits for the Lead's review, and availability tells idle apart from awaiting-review or a failed last turn. With delegationEngine=auto it uses independent sessions, continuable subagents, or the kernel roster depending on what the composition provides.

DeepSeek Harness plugin that catches the wrong package manager for the repo (npm install in a pnpm repo, pip install in a uv/poetry project, and so on) and gives the model the exact right command.

Keeps heavy tool groups out of every request, and can scope tools per session or per role: the role key comes from the host's own session label, subagent marker or preset, so an orchestrator's child agents each see only their own group allow-list. Also auto-groups new tool families by name prefix, lets the agent classify ungrouped tools once, brings a hidden tool back when it is actually called, and exposes a health snapshot of what each session can really see.

Homepage-style chat background skin for DeepSeek Harness, ported line by line from deepseek.com/harness: WebGL fluid background (5-colour noise field, flow-field brush, light, bloom, vignette, grain), white 90px dot-line grid with mouse perturbation and spring return, and a digital point-cloud whale that assembles from particles, breathes, swims, is pushed radially by the cursor and brightened by a light beam. Dark theme only; disabled automatically under reduced-motion.

A chip at the left end of the composer tool row showing whether DeepSeek bills at peak or off-peak right now. Clicking it opens a card with the current state and rate, the next switch with a countdown, and why this instant is in that state: a weekend, a named statutory holiday, a 调休 make-up workday, or simply outside the peak windows. Times are resolved in Beijing time with an explicit UTC offset, so the verdict does not depend on the machine clock; a hard-coded 2026 Chinese holiday table carries the 33 holiday days and the six 调休 workdays, and a year beyond the table says so instead of guessing.

Prompt optimization, next-sentence prediction, inline completion, and per-session draft stashing for the DSH input box.

A long conversation that renders many LaTeX formulas at once can become laggy, because the browser keeps styling, laying out and painting content that is off screen. This plugin applies content-visibility: auto to the off-screen message bodies, so that work is skipped and the lag is reduced.

A Versions tab on the Plugins page: one-click upgrades that replace the plugin in place where they can (no uninstall, enabled state preserved, falling back to remove + reinstall with the reason logged when the profile writer lock is busy), live pnpm output with per-step timing and cancel, the full published version history with one-click rollback, every version checked against the running DSH so incompatible ones are marked and blocked, a rollback that installs the old version but cannot activate it reported separately from a lost package, profile manifest writes taken under the harness writer lock with one recoverable journal record per package after a failed upgrade, mirror-gap warnings, and a selectable npm registry saved in the profile.

Use DSH from a phone browser. Turn on phone access from the phone button in the session header (or the plugin card) and scan the QR code; the phone opens the official DSH web UI served by the desktop process, with the same sessions updating live, so you can follow a running task, reply and approve away from the computer. A gateway inside DSH (port 18765) proxies the web UI and handles the desktop login itself. One QR code works on Wi-Fi and on mobile data: turning access on also starts a free Cloudflare quick tunnel (cloudflared is downloaded from Cloudflare's GitHub releases if missing), and its address survives DSH restarts, so phones do not need to rescan. Pairing codes are one-time and expire in 10 minutes, each phone gets its own revocable credential, the header button shows when a phone is connected, and idle connections get keepalive pings. Optional push notifications go to Bark, ntfy, ServerChan, Feishu / WeCom / DingTalk group bots, Telegram or any webhook when a task finishes (after 30 seconds or more), stops with an error, or waits for an approval or an answer; they carry only the session title and status, and approval still happens in DSH. On a phone the web UI's side panels slide over the page, and it can be added to the home screen with its own icon.

Android development toolchain inside DeepSeek Harness: 23 android_* tools that find the SDK, create and manage AVDs, start and stop emulators, scaffold a Kotlin/Compose project, build and run it with Gradle, then install, launch, screenshot, tail logcat and drive the UI. A settings page holds the SDK, emulator, API-level, build and output parameters.

Agent-facing plugin installer that resolves DSH-compatible versions, handles ambiguous names, and verifies the loader entry after install.

Windows desktop notifications for DSH that speak the turn result by default: Windows TTS needs no download, and 13 VOICEVOX voice packs can be installed from the settings page. A completed alert waits until running subagents and background jobs finish; approval, question and error alerts fire immediately. Voice packs are downloaded from GitHub after a terms confirmation.

Windows toast notifications for approval requests, questions, completed or failed runs, and sign-in prompts; clicking a notification reopens the conversation that raised it, and toasts are attributed to the DeepSeek Harness app identity.

Registers native doubao_generate_image and doubao_image_status Agent tools to automate Doubao Web (doubao.com) via Tabbit Browser, automatically extracting watermark-free 2.7K wm0 original images into local files and multimodal chat attachments.