Plugins

Entrada de voz mediante la Web Speech API del navegador: sin servidor, sin claves, sin descargas de modelos (Edge = Azure, Chrome = Google speech).

Perfiles de prompt en Markdown reutilizables para DeepSeek Harness, con selección de modelo por turno, sustitución de argumentos y restauración de estado.

Backend de telemetría que exporta turnos, pasos del modelo y llamadas a herramientas a yiTrace.

Memoria entre sesiones para el modelo: búsqueda de texto completo en todas las sesiones pasadas (SQLite FTS5 vía ctx.sessionQuery) y recupera los fragmentos con mejor coincidencia hacia el contexto actual.

Many model sources behind one authenticated OpenAI-compatible endpoint for DeepSeek Harness (dsh) and your tools: WorkBuddy, OpenCode Zen and your CLI subscriptions (Codex, Kimi…), with a management page inside dsh. 多家模型汇成一个 OpenAI 兼容地址，dsh 和其他软件都能直接用。

Restores Host-persisted settings when dsh web is reached through a reverse proxy on a non-loopback hostname: marks the page connection as host-owning so the Settings and model pages load instead of reporting "settings are unavailable in this browser". For dsh 0.1.7 – 0.2.x.

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.

Read-only Git and SVN code review using the current DSH agent model, with change previews, explicit context and approval-gated model requests.

Dependency-ordered multi-agent team runner for DeepSeek Harness: define roles with their own model, persona and tools, run a node graph with an independent review gate and bounded rework, and watch progress in the web UI.

Lets Codex delegate tasks to Desktop sessions through a local authenticated bridge and companion MCP server, with Workspace grouping, model and reasoning effort selection, event tracking, waiting, and cancellation.

Reads the session event stream, with no extra model calls, and turns corrections, failures and fixes into pending lesson candidates. Every candidate is judged against written criteria and carries the reason it passed or was refused; only confirmation writes a skill, and only through the single validated writer `learn_skill_manage`. An idle curator marks unused skills stale and archives them by moving rather than deleting, and spares any skill that is still being loaded. Five tools: `learn`, `learn_review`, `learn_curator`, `learn_skill_manage`, `learn_skills`. Registers a prompt section stating the criteria, and a tool guard that refuses direct edits to skill files so the writer cannot be bypassed.

Keenable web search and page fetch for the built-in web_search and web_fetch tools, working with any model and no API key.

Delegates self-contained coding subtasks to the local Claude Code subscription through the official Agent SDK and returns the result to the host session. Nineteen per-call parameters control model, thinking effort, permission mode, allowed tools, turn count and budget cap; background jobs return a jobId immediately and stream incremental output; the result carries a session handle that can be passed back as resume to continue the same conversation; commands expected to run past ~20 seconds go through the built-in mcp__dsh__shell channel, which streams output line by line without consuming model context; a Claude Code monitor panel in the session lists jobs, renders the structured process, shows live output blocks and cancels a run; the delegation procedure ships as a runtime-registered skill, so installing the plugin installs the skill.

Stops DeepSeek Harness from compacting (summarizing) your conversation far too early. Without it, compaction fires around half-full on smaller context windows; with it, compaction waits until the window is as full as you want (80% by default), settable per provider/model. It only changes when compaction happens — what a summary keeps is still DSH’s decision.

DSH plugin: hands a long session over to a fresh session in the same workspace on /handoff or at a token threshold, carrying the preset, model, permissions and plan mode across; the brief is written by the session's own model and the switch waits for a step boundary instead of interrupting running work.

Uses a dedicated model for context compaction rescue and adds manual compaction to the native context popup.

Adds configurable reasoning levels for third-party models and a model-strength picker in the composer.

Edit reasoning effort levels and thinking formats for models that declare none.

A Windows desktop pet with optional quiet mode, classic/gentle feedback and independent reduced-motion/flash/floating-text controls, preserving classic defaults and updating real data without replaying old events. Includes five independent cardTheme palettes (default/original blue, forest/green, amber/brown, violet/purple-gray and paper/white) freely paired with six 2D character skins; each palette supports DeepSeek wallet, peak/off-peak pricing and cache information, or Codex 5-hour/weekly quotas and reset times. The host-native DSH settings gallery shows both modes and character/card combinations with clearly labeled synthetic preview values; explicit saves hot-reload without new model requests or account changes. v0.6.5 separates disk saves and drafts from a recent successful native application receipt, verified against the owned process/run/profile, raw revision, effective preferences and a short lease; stale or failed evidence never claims success. Retains draggable positioning, inert right-click, four sizes, preferences/statistics, responsive light/dark settings, authenticated atomic saves, configurable quota polling and default-off deduplicated warm-up.

Composer prompt polisher: a button in the input tool row rewrites the current draft through the session model (workflow engine, enabled automatically) or an OpenAI-compatible standalone API with host-side key storage, backfills the rewritten text with a replace/give-up dialog when the draft changed mid-run, and ships concise/structured style presets, custom templates with a {draft} placeholder, one-click restore from the status dock and reference-chip protection.

Suggests the next user reply in an empty composer after the assistant finishes, using the session model and rolling context summaries, with Tab to accept.

Localizes the DeepSeek Harness approval panel in the web client and translates authorization reasons from English to Chinese, with a built-in phrase dictionary first and a model-router fallback for reasons the dictionary does not cover.

Interactive terminal UI for DeepSeek Harness: tables, Mermaid and chart rendering, side conversations, and the same Cordis plugin model as dsh Web.

Composer-dock badge showing DeepSeek peak/off-peak pricing on a Beijing-time calendar whose China holiday schedule is fetched online at runtime (off-peak costs half the peak rate); auto-hides for non-DeepSeek models and cycles seven readouts on click.