Плагины

Перетаскиваемый талисман приглашает сыграть в сянци (китайские шахматы), пока ИИ думает, со встроенным движком, экспортом записи ходов и опциональной мультимодельной комментаторской поддержкой.

Переписывает каждый запрос согласно выученному профилю промптов и автоматически обновляет привычки, правила против галлюцинаций и записи обучения.

Плавающий предпросмотр в стиле Google AI Studio для веб-интерфейса: перетаскиваемый и изменяемый по размеру iframe-предпросмотр с пресетами соотношения сторон и разрешения, плюс окно кода в стиле VS Code с прямым редактированием, подтверждаемой записью и записью в журнал сессии.

Удаление сессий из работающего веб-runtime: живое хранилище, записи рабочего пространства и артефакты на диске (без перезапуска).

护理记录时限、签名与评估表完整性校验（依据《病历书写基本规范》等公开文件，仅提示差异，不作出定性结论）

Session-scoped delivery receipts for local artifacts: compare current file hashes with a recorded baseline, keep agent claims separate from explicit user acceptance, flag changed or unavailable files, and export a Markdown receipt through a tool, slash command, and conversation panel.

Mount the QianYuan MCP server inside DeepSeek Harness: one config-layer bundle brings 15 agent-record tools (pitfalls, cached results, identity, task board) to mcp__qianyuan__*. No runtime code; reads are anonymous, writes are ed25519-signed.

Tag each session as a long-term or short-term task, record the agent mode it was created with, and auto-archive it once the archive window you set elapses.

Select text in the chat to quote it into the composer as an annotation, or to ask about it in a throwaway side panel that leaves no record.

Turn a raw recording into a published podcast with a local ffmpeg — detect chapter boundaries from silence, cut at those boundaries, normalise loudness to a broadcast target in two passes, and write an iTunes-compatible RSS feed whose enclosure byte lengths are measured from disk.

Cross-session persona and diary for DeepSeek Harness: a local gate decides whether a turn is worth remembering, then one extra model call turns it into a first-person diary entry plus an impression of the user, where every impression carries the specific moment that produced it. Later conversations inject about 2000 tokens of that as a self-memory block in three registers: a persona core, a section for judgements the model wrote itself, and the diary. There is a tool for the model to record a judgement in its own words, and a plain markdown hand-note file the human can write into which the plugin only ever reads. A sidebar panel lets you read, forget or export the archive, and forgetting demotes an entry rather than deleting it, so an old one still surfaces on a strong topical match. Local-only, secrets are stripped before anything is sent, and the archive is a plain folder you can read, edit or carry to another account. Zero runtime dependencies, no build step.

Visual frontend revision workbench: point at a real page element, describe the change, send it into the current DSH session, then review fixed-viewport before/after screenshots (1280x800 and 390x844) beside a captured source diff, accepting each revision item by item. Revision records are kept per session under .dsh-revise/ and export as a single portable HTML report.

Restarts the running DSH process from a model tool or a bottom-right GUI control, with launchd KeepAlive detection, a detached-supervisor fallback that aborts the exit when its handshake fails, a macOS desktop restart path, and JSONL restart records correlated by arm id, pid and profile so a restart stays distinguishable from unrelated log activity.

It shows the number of confirmed loaded skills in the conversation header. Click **技能** (Skills) to open the right sidebar and check whether each skill's full main body remains in the current main assistant context, along with its recorded versions and sources.

Record every step the agent took, and walk through it forwards and backwards to find where it went wrong.

Every change to your agent's instructions is recorded with a diff, and any change can be undone.

One-click publish of DSH conversations as Markdown articles on a Halo CMS blog: confirm in the GUI, type the console password (never stored), and the Q&A record is created through the Halo REST API.

Pure JavaScript macOS Keychain credentials provider for DeepSeek Harness, replacing plaintext .credentials.yaml storage and fully supporting both refs and records without native builds, signing, or Xcode.

Memory bridge between DeepSeek Harness and ai-memory: runs ai-memory's finalize-session when a DSH session ends, so the observations it already captured become searchable wiki pages. Skips subagents, queues concurrent session ends, and sweeps sessions abandoned by a forced restart. Requires a separately installed ai-memory server with its DSH lifecycle hooks wired by hand — ai-memory’s own install-hooks does not support DSH; this plugin only finalizes what capture has already recorded.

Connects DeepSeek Harness to the QianYuan MCP server, a shared record of pitfalls other agents already hit and results they already produced, plus capability lookup. Exposes 13 tools named mcp__qianyuan__*. Reads are open and need no key.

Speaks a short status line when DSH changes state — turn done, turn error, job done, job failed, goal complete, goal blocked, an approval waiting, or the agent needs your answer — using eight pre-recorded clips in Chinese (default) and English. The audio is played by the host process through ffplay, or through the Windows PowerShell player that ships with the OS, so it does not depend on a Web Audio tab. Windows only.

DSH plugin for bounded agent evaluation: compare an original with candidate prompt or supported configuration changes against user-supplied checks, record evidence and costs, and leave adoption to the user.

Agent screen capture on macOS and Windows: one tool captures the screen and returns the image itself, so the model sees it without a second call. On Windows a whole burst runs in one engine call; on macOS a resident helper drops a region capture to 13ms and a change check to 23ms. A second tool, macOS only, reports whether Screen Recording is granted and opens the settings pane that fixes it.

Computer Use for macOS desktop applications: reads any app's accessibility tree as text with element indices, then clicks, types, pastes, scrolls, drags, selects text and sets values through it, plus screenshots read back as images. Driven by a persistent JavaScript REPL — the agent gets two tools (`js`, `js_reset`) and calls an `sky` API inside them, so state survives between calls and the tool schema stays fixed as capabilities grow. Accessibility trees come back as diffs after the first read; actions address elements by index and indices from a stale read are refused rather than misapplied; typing waits for keyboard focus and reads the field back; paste restores the clipboard. Requires macOS 14.4+, Accessibility permission (Screen Recording only for screenshots) and Node 20+; ships a prebuilt native module, so nothing compiles at install time.