Plugins

Design I/O for coding agents: controllable UI generation via declarations (spec/domain/craft/design/components/template) and contracts (skill/evaluator). Use for product UI—console, dashboard, agent-ops, CJK-first apps.

Reusable Markdown prompt profiles for DeepSeek Harness with per-turn model selection, argument substitution, and state restoration.

File-based memory for DSH agents: layers, FSRS-5 decay, cold archive, recall, zones, and optional auto-injection.

LAN access gateway for the Web UI: serve DSH on localhost or 0.0.0.0 with password protection or an explicit no-password mode plus expiry presets, and open a separate chat-only port that other agents can drive over HTTP with selectable workspace, token or token-free auth, and its own expiry. LAN sharing is opt-in: it rebinds the Web UI port to 0.0.0.0, gives other devices the full DSH interface behind a password (the default), and an explicit no-password mode needs you to confirm a security warning.

Adapts Easy Ask AI Skill's Prompt Ready and Get Clear flows to Harness chats with a native sidebar; the original Skill can be installed in other compatible Agents.

Adds a "Delete session" action after "Archive session" that permanently removes one session: its log directory, rewind snapshots and projection-cache entry, its workspace/archive/pin registrations, schedules bound to it, and its rows in the chat-index and usage databases. A preview lists every target before anything is deleted; the same operation is exposed to agents as the session_delete tool with a dryRun option.

Bind prompts to individual providers and models for use across Standard-mode conversations and subagents.

AgentScope Platform inside DeepSeek Harness: manage several platform accounts (switch the billing account by hand, pick one by remaining Credits, probe and flag unhealthy ones), an optional OpenAI-compatible model provider whose catalog you can fetch, add to and edit row by row, an available-Credits and redeemable-contribution readout under the composer, and a settings panel for the daily check-in, the balance, contribution grants, usage detail and the price list. Switching accounts affects new requests only and never interrupts a session already in flight.

A Dynamic-Island-style status panel that lives in the macOS notch and lists running turns, background jobs and subagents, with a completion banner and a spoken reminder.

Drive DeepSeek Harness from WeChat: scan a QR code to sign in, send instructions in WeChat, and the agent reply lands back in the same chat window. Several agents can share one WeChat account, each with its own model, preset and context, switched from the chat with /use. Only the configured owner WeChat account can drive the agent; other senders are rejected by default and require explicit approval.

Batches independent tool calls into one model round-trip: run_tools dispatches many steps concurrently (after-dependencies for ordering), run_parallel fans out sub-agents, so fewer round-trips and lower wall-clock time.

Kaipu Workstation — a cross-cloud, cross-platform console where agents collaborate on one workspace. Any agent runtime that speaks plain HTTP can take a seat, each with its own device credentials: project context is shared, files are kept per seat, and tasks rotate between seats with a trace of who did what. Workstation data lives on the server — this plugin is only the interface, so uninstalling loses no data and reinstalling restores it. The panel renders the persisted workstation state (no live stream yet). Workstation data is stored on the author's hosted server (agent.ak2029.com/kai-pu) unless you override baseUrl in the profile config.

An audit workspace for DeepSeek Harness: a multi-agent consultation venue where each review seat is filled by an agent, every judgment is routed to a named review face, and the whole run is kept as a traceable record. Works out of the box: the service endpoint is built in, a workspace is provisioned automatically on first launch, and review scenes come with their seats and agents listed. The built-in service endpoint is the author's hosted server (agent.ak2029.com/kai-pu): scenes, runs and review records are stored there unless you override baseUrl in the profile config.

Syncs DSH skills, profile settings and AGENTS.md with a private git repository: status chip, six-section health check, pull, secret redaction, and a snapshot-then-push flow. Dry-run by default, never rebases or force-pushes.

Tool-surface governance for DeepSeek Harness: six scope-separated deny lists cut each agent's tool visibility through the host tools.restrict() API, unseen tool names wait in a pending gate, and dispatch_composed assembles a minimal surface from capability tags — a tasks array runs subagents in parallel, and a continued subagent keeps its original surface. Every decision goes to an append-only ledger that rotates into dated archives without ever losing history.

Long-term memory ported from the OpenAI Codex memory pipeline: sessions idle for 6 hours are extracted in the background by a chosen model, then merged by a hidden agent into a Markdown memory folder at most every 6 hours; new agents get the memory summary in the system prompt plus four tools to list, read and search memory and save corrections. Models are chosen on a settings page or with /memories; an optional filter keeps a notes folder out of memory.

Defines named roles that each pair a persona with a bound model and fallback chain, and exposes one delegation tool per role, so subagents are dispatched by role name instead of a provider/model string.

Brings Google Antigravity (agy CLI) models into DSH with streaming, thinking and native tool cards, a manual multi-account pool with quota and proxy controls, and an optional experimental direct transport. Community fork of amlyczz/dsh-agy-link. Google's Antigravity FAQ says personal plans do not support wrapping agy in third-party coding agents, so using it may put your Google account at risk of restriction. Needs Node.js 24 or newer and Google's agy CLI on PATH; the experimental direct mode is off by default.

Grok-bot-style AI organization in the sidebar: a chief-of-staff agent creates departments, roles and per-department group chats, dispatches each task to child agents running on this machine, and reports finished work back as one assistant message. Its HTTP routes have no DSH access check, and they can create agents and dispatch tasks to them.

A Windows Harness adaptation of [Media Review Skill](https://github.com/Ww-Cooooo/media-review) for AI video feedback: annotate frame regions or time ranges and jump to recognized speech or imported subtitle cues; other Agents with local file and command access can use the original Skill. Windows only and built for DSH 0.2.2-rc.1 (a 0.1.0 release candidate). Waveforms use an existing FFmpeg/ffprobe and transcripts an existing Whisper; without them, manual annotation and subtitle import still work.

Adds one system-prompt section that limits delegation to subagents and teammates to two cases: adversarial review of freshly produced output, and one-shot context reading that returns a summary.

Writes engineering decisions down the way DSH documents its own — every non-trivial change leaves a Markdown note under .agents/notes/<lifecycle>/<class>/ holding the problem, the decision, the rejected alternatives and the consequences, committed to git and reviewable like code. It supplies the workflow for that: five tools (note_search, note_read, note_write, note_verify, note_skip), an end-of-turn hook that asks a code-changing turn to either write a note or record an explicit skip, the tree/format/archived checks, and the write-notes-like-deepseek skill bundled in the package and mounted by the plugin. A sidebar board that renders the notes into four views (architecture baseline, timeline, rejected proposals, decision list) comes with it; the notes themselves stay plain Markdown.

Project workbench for DSH: groups sessions by project (workspace) in a panel with project cards and session lists, applies a project's default preset, model and reasoning effort to new sessions, scopes memory per project on request, previews AGENTS.md/CLAUDE.md read-only, and groups the sidebar by project at startup.

A cached daily scene, real weather, schedule and outfit for dsh conversation agents, with three ootd_* tools and the /ootd command.