Plugins

macOS only: holds a PreventUserIdleSystemSleep assertion (caffeinate -i -w <host pid>) while any DSH session runs, and releases it when the last one stops, so a long turn is not cut off by idle sleep. The display may still turn off. It keeps holding through the maintenance phase, where Agent.status reports idle and a long manual /compact runs, via the compaction/start and compaction/end events. It reconciles against ctx.agents.list() every 60s, so a session already running when the plugin is hot-reloaded is picked up at once instead of at the next turn boundary. A failed hold retries with exponential backoff capped at 60s and never gives up; release escalates from SIGTERM to SIGKILL after 2s, and -w lets caffeinate exit by itself if the host dies. No browser half — configured through the profile patch. 51 tests, including two that read pmset -g assertions on real hardware.

Native macOS completion notifications and user approval alerts with Settings UI for DeepSeek Harness.

Recurring Excel template fill and reconciliation review: declare the file, sheet, header row, match key and amount columns once, preview every cell write before it lands, compare amounts as exact two-decimal values, list the unmatched rows as exceptions, and export a filled template copy with an HTML report and a JSON trace back to the source rows.

Declares a spec-driven development agent preset with a Chinese persona and four static Claude Code/Codex delegation tools for planning, implementation, review and knowledge distillation; roles are prompt-guided, not an automatic state machine.

Read, audit and diff Protocol Buffer definitions without installing protoc: a zero-dependency .proto parser exposing six tools (status, outline, describe, diff, audit, next_number). diff classifies every change as breaking, risky or additive and names the wire-format consequence; audit catches what protoc accepts but a reviewer should not — duplicated or illegal field numbers, reserved-range collisions, unresolved types, and unreserved gaps below the high-water mark.

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.

Drive a local ffmpeg as six structured tools: probe a file's streams, transcode or trim it, pull evenly-spaced frames or the audio track, and compress video to a byte budget with two-pass encoding.

Pack a directory into a deterministic tar, read an archive's contents without extracting it, unpack while refusing every member that could reach outside the destination, and verify what came out against what went in by SHA-256. No external dependency — the tar format is written and read in this plugin. Extraction is deliberately more permissive on reading than on writing, because a reader's job is to report what an archive says, while a writer must never emit what it cannot read back. A tar has no central index, so the walk tracks whether it reached the end-of-archive marker: without that, a half-file checksums fine and reads as complete.

Per-turn development progress memory: completed turns are classified into date-indexed Markdown files under doc/progress/ with a maintained INDEX.md, system-level bugs are kept in a cross-project SYSTEM_MEMORY.md and re-injected on error keywords, and a progress_read tool loads full documents on demand.

Bounded self-managed long-term memory (Hermes-style MEMORY.md / USER.md with hard character limits and a safety scan), a review that runs after every turn and writes plain-language notes or refines an existing skill instead of adding a near-duplicate, a curator that finds and surgically repairs skills the loader would silently discard, cross-session search over past messages, and a learning card above the composer that appears only when there is news and asks before it deletes or rewords a memory.

Compaction backend that replaces the default compaction-basic: it summarizes the conversation with a configurable pool of low-cost models, sizes each chunk to the chosen model's context window, and merges all digests in one call, with per-route cooldown and session-model fallback.

Xiao character theme for DeepSeek Harness Web and Desktop with light and dark palettes, selectable illustrations, a draggable companion, session-state feedback, and optional official DeepSeek balance display.

Aqua glassmorphism theme for the DSH Web UI, adapted to DSH v0.2.x and desktop v0.22.3+: the header, sidebar, composer, stats line and trajectory view become frosted glass panes over a live WebGL fluid board or an image/video wallpaper, with mica and compatibility layouts, adjustable blur, frost, hue and brightness, marine decorations (particle whale, fish, mesh, cursor glow), a one-click reset-to-defaults button, and a master switch whose off state restores the stock UI exactly; adapted from WYH66666666's original (0.1.x) and du-u-uck's 0.1.1-rc.2 port.

Rewrites the draft in the DSH composer into a clearer, more executable prompt through the harness's own LLM routes, showing an original/optimized comparison before you adopt it.

Swaps the composer's model control for a PI-Desktop style picker: a two-pane menu with search and provider grouping, a reasoning-effort slider, ArrowUp/ArrowDown prompt-history recall, and a prompt-rewrite button with undo.

A read-only Cordis Inspect provider that lets an agent see the interface it is driving: outline() reports which UI regions the running shell actually rendered, with occupancy, geometry and current or past failure state; snapshot() returns one region's rendered node tree (tag, role, aria-label, own text, geometry, computed styles) addressed by slot key or CSS selector; failures() keeps a history of regions that failed to render and whether they recovered. Regions are addressed by the anchor the shell already writes for every slot, so it needs no region list and covers regions added later or contributed by other plugins. No UI, no tools, no dependencies, no network access.

Smart prompt optimization & dynamic thinking effort adaptation with composer capsule, DSH/custom models, undo/re-optimize, and native settings.

Shows what each completed turn cost in CNY on its own row under the turn, with a click-to-open breakdown naming the tokens per category, the unit rate, the matched price rule and the billing instant; per-model rates are edited on a settings page inside DSH — four token categories plus weekly recurring peak/valley rules with arbitrary multipliers and absolute time-window promo prices that override them. 0.2.3 adds account awareness: it reads the balance or remaining quota of 14 providers (DeepSeek, OpenRouter, Zhipu GLM, Moonshot, MiniMax, Kimi Coding, OpenAI-compatible relays and more) and shows today / this month / this session spend next to it; spend is accumulated in a local ledger (atomic writes, one book per API key, 1-24 month retention, JSON export), and a monthly calendar lets any day be corrected by hand - including days with no balance observation - with an optional note. Usage and rates are computed locally from what the provider reported; balance is read straight from the provider API with your own key, and nothing is sent to a third party.

Message recall for DSH: cut the conversation back to before a chosen user message so it leaves the model's context, hide it in the transcript, and put its original text back in the composer. Works through the kernel's own surface-replacement mechanism, so

Evolver for DeepSeek Harness — persistent, auditable evolution memory: recalls what worked behind each prompt, detects improvement signals while you edit, reports each reuse back with graded evidence, and exposes the EvoMap network's genes and capsules as

Recovers broken DSH turns: widens the product's own retry budget and renders the wait as a live countdown inside the same step, keeps a truncated output alive inside the same turn, and can open an opt-in continuation round when a turn has already ended; while the composer is empty it hands the send button to that continuation.

Buffers consecutive short messages from the composer and merges them into one turn: no model wake-up until the user stops for a configurable silence window; long messages and attachments bypass the buffer.

Prices tokens with the official model price table, keeps the per-session and global running cost beside the composer context meter, mirrors your DeepSeek account balance in the sidebar, blocks a turn once this period passes your own global ceiling (clamped to that balance) or a per-session ceiling (each session sets its own, independently), and resets the billing baseline per session or globally — the two are independent, so resetting one never resets the other — so the period goes back to zero while history only rolls into the lifetime total.

Returns the caret to the session composer on Ctrl+I; the key is rebindable in Settings and works from any region of the desktop window.