Plugins

Desktop notifications for DSH: a system toast, a per-kind chime and a tab-title marker when the agent needs you (approval, plan review, question) or a reply finishes in the background; errors, interrupts, usage limits and multi-agent workflow progress alert too. A host half adds SSE delivery, webhook push (WeCom/Feishu/DingTalk/generic), per-turn token/cache/TPS stats, plus three message-format presets (concise, detailed, custom); settings changes auto-save.

Recall a message you already sent: the session log stays append-only and nothing is deleted, while the recalled turns leave the model context and the transcript, the text returns to the composer, and the files those turns changed wait in a quarantine you can review and clear from its own settings page.

Registers a real-browser URL-to-text MCP server in a DSH profile: fetch_url and fetch_urls render JavaScript-heavy and anti-bot Chinese pages (Zhihu, CSDN, Juejin, cnblogs) and return clean article text as JSON with per-tier diagnostics, and crawler_status reports whether the browser is available. Needs a one-time Python dependency setup after install.

Deep-sea skin for the DSH Web UI: an 80-token light/dark theme layer, a two-layer background, a gradient sidebar with a whale sticker, a wordmark that doubles as the skin switch, and a rotating turn-status line.

Circuit breaker for idle agent turns: watches agent/assistant-stream and cancels a turn that produces only reasoning — no text and no tool call — after 45s, after 80000 reasoning characters, or on a degenerate repetition pattern (exact repeat unit, sentence-template repeat, sparse repeat, or 6-gram density). On trip it injects a continue instruction that requires the next action to be a tool call, and ships 11 regression cases plus per-gate tuning.

Edit any user turn in DeepSeek Harness and re-run it: an inline editor on the message row rolls the conversation back to just before that turn through the official surface-replace contract, then re-prompts the agent with the revised text. The append-only

Active wait for background subagents in DeepSeek Harness: registers a wait_subagent model tool that blocks until a named background continuable subagent settles, then returns its stop reason and closing message — optional timeout, membership gate rejects unknown ids, event-driven with zero polling. Closes the gap between run_in_background fire-and-return and the async settlement notice; wait for a background subagent without polling list_agents.

On-demand skill tools for DeepSeek Harness: registers skill_search / skill_load / skill_ref so the model searches a staged skill library first and loads a skill body only when it needs it, keeping the library out of the resident session catalog and out of every turn's context; adds a Skills tab to the conversation view ring that reads back the whole session and lists which skills were actually loaded, and in which turn. Zero dependencies, and the records render client-side only, never entering the model's context.

Mounts the mcptoon MCP gateway as one stdio server, so the agent discovers the tools of every MCP server you have configured through mcptoon's compressed view instead of full JSON schemas on each turn. Ships a doctor CLI that checks the mcptoon executable resolves and prints the exact command/args override when it does not.

A dsh-better-sidebar tab that renders a session brief — a todo board read from the `todos` projection plus a deliverables section listing the files written this turn — and hides the official todo panel above the composer while it is mounted.

Turn event notifications: six event classes (completed, error, interrupted, approval requested, AI question, limit reached) delivered via sound, system notification, in-page banner, webhook, IM and host desktop-notification channels; when several browser windows are open, only one rings.

Session usage dashboard: day/hour/minute token and request statistics, a 26-week activity heatmap, cache-hit-rate curve, per-model donut and list, and a settings page; it appends its cost item after the official session bottom bar (per-turn cost chips included, no separate detail button), renders the official cache-hit-rate number with two decimals, and estimates costs from multi-condition, time-segmented pricing rules. Re-implemented from HaoyueQin/dsh-usage-statistics-panel with extensions.

Conversation enhancements: a history-input overlay (Alt+Up, with workspace/session/global/favorite-prompt scopes), recall of interjected messages that were not applied, rewrite-style forking (fork a child session from before a turn and backfill that turn input for editing, including turns whose reply is still running), optional auto-resend after forking (off by default), and a copy-session-id action in the title bar.

Turns a plain-browser DSH session into the WhaleHarbor lite client: a standalone browser --app window with a client-style top bar and task notifications, while the full desktop installer downloads in the background (multi-source, SHA-256 verified) and an install prompt appears when ready.

Native desktop notifications for DSH — approvals, questions, turn completion, session end and errors — with click-to-session, per-event toggles, quiet hours, a privacy switch that keeps message text off the banner, and a Web Audio tone per event type.

DeepSeek Harness plugin that keeps the agent's green claims honest: verification runs become receipts, edits are tracked, and a stale, failed or masked "all tests pass" is turned into a rerun instead of an ending.

One-click Concise output style for DeepSeek Harness: results first, no filler, same work depth. Every final reply opens with a plain-language digest card (select text to copy); file paths inside the card are highlighted and click-to-open with the OS default app, and a deliverables zone lists the files the turn actually produced from real tool outputs. Scoped per session, works in headless/CLI too, with /concise command, style.md overrides and a local API.

Resumes an agent turn that ended after narrating its next action without calling a tool: the guard replays the turn log on agent/turn-stopping, asks a judge model one true/false question, and steers the same turn to run one more step.

Notifies you with a browser notification and toast whenever a session turn ends, with 5-language templates and custom presets.

Streams DeepSeek Harness sessions, messages, tool activity, and turn outcomes to a JARVIS Mission Control board with task status and human review.

Hand a DSH web session to the locally installed Claude Code CLI: a per-session engine selector (DSH | Claude Code), Claude models and reasoning effort in the model seat, Claude permission modes in place of the access-mode selector, and permission requests, AskUserQuestion and ExitPlanMode routed to the native DSH approval, question and plan-review cards. The stream is written as native DSH session events, so the built-in transcript, tool cards and nested subagents render it; Claude slash commands merge into the command palette; the CLI is held by a detached broker, so plugin updates and DSH restarts do not interrupt a running turn.

DeepSeek account balance, this session's billed spend, and today's total spend in the session-header badge; a turn-cost row in the message actions strip, a detail panel with per-model pricing and a today session-spend ranking. Official peak/off-peak pricing; balance from the real /user/balance API.

Real-time usage, cost and balance meter for DeepSeek Harness: official-API balance, cost accuracy within ¥0.02, per-turn / per-session / per-model cost ledger, live color-tiered token speed and a docked one-click popup.

Separate slash-command group in the Web UI composer: /steer injects text on the agent's next step without interrupting the turn, /new starts a blank session, /skill and /docs wrap /steer with post-turn skill or docs tasks, and custom /steer aliases are configurable in Settings.