Plugins

Two-level model routing: pick the best model group first, then the best channel within that model; a failure switches channel before switching model. Preference mode pins one model across its channels, a status bar under the composer shows the live model, channel health and the day's token usage, and a speed/intelligence report can be imported as the routing prior. Exposes a model_orchestrator tool and a /route command.

Registers local Qoder, QoderCN, OpenCode, Pi and CodeBuddy CLIs as DSH subagent backends, with ordered model fallback, live progress, per-run credit display, a settings page, and slash commands. Requires the corresponding CLI installed and logged in.

DeepSeek Harness plugin that trims tool output (head/tail plus grep folding), adds optional prompt shaping (Caveman and Ponytail), instant local responses, RTK command-output compaction, context-pressure handling, optional LLM-summary compaction, stable prompt-cache layout, a Web settings UI, and persisted savings telemetry.

Adds a gear button beside the sidebar account row that opens the Settings panel in one click. It runs the host's own settings.open command, so it reaches the shell store directly instead of going through the account menu, and the behaviour is identical on desktop and on a phone over the LAN where no keyboard shortcut is available.

A conversation-style layer that rephrases trade-off warnings: when a shortcut is cheap but has a catch, the reply may say the fast road is the rival shop and the slow one keeps its rules. Ships a rewrite engine that turns any text into a fixed-format dialogue with five independent knobs (act count, catchphrase density, ending turn, seed, preset), a slider for four on/off intensity levels, and a /bushaoxin command that reports a joke "heartburn index" from the session tool-call log. Every rewrite ends with a real technical conclusion that cannot be disabled. Zero dependencies, no build step, engine is pure functions.

Phone-side command center rather than a web-UI mirror: dispatch tasks into isolated git worktrees from a PWA, tool-approval push notifications with phone/desktop first-answer-wins racing (never auto-approved), a multi-machine session board, and an end-to-end encrypted zero-knowledge relay you can self-host.

Publishes an ordered system prompt section that names wsl as the command executor to reach for first, so the agent prefers WSL over PowerShell on Windows. Registers no tools, and only advertises executors the agent can actually call.

Bridge to the UnrealAgentLink editor plugin: the Unreal Engine editor dials this WebSocket bridge on 127.0.0.1:17860 and the agent drives it through 12 tools covering all 170 cataloged commands (actors, levels, blueprints, materials, PCG, screenshots, logs, undo).

Compiles an LLM persona in DSH from six dimensions (identity, tone, creative habits, feedback style, boundaries, run parameters), injects the compiled text into the system prompt at the end, ships six built-in templates plus a drop-in templates folder, and drafts a persona from your own past messages — either by counting patterns in them or by having the current model read them and report its rationale. Adds a panel next to the composer and a /persona command.

Read-only clinic for DSH session logs: 22 named diagnostics (token waste, retry and failure loops, context bloat, cross-session amnesia, idle MCP servers and skills), each reported with a replayable evidence locator such as session.v4.jsonl.zstd#L101-L110, and each finding turned into an executable prescription — a profile-patch YAML snippet or a command with current to suggested values, risk, verification and source — plus a feedback command that retires a prescription you reject, after which the report states it was removed by your feedback. Zero dependencies, default zero egress, session files are never written.

Blocks AI coding agents from editing files without evidence and from running unapproved high-risk commands, and tells quoted dangerous strings apart from commands actually being executed.

Hand a command-line-verified operation over to the GUI: drive it once, verify with ground truth, and hand back a self-contained illustrated tutorial.

Injects the exact current local time and the timestamp, day relation and recency gap of the previous user message into every prompt, and adds a /time command that prints the injected block.

Deterministic goal-completion gate for DSH: a contract-first AC grammar with named failable check commands, a gate that re-runs every check itself (zero model trust), verdicts bound to a tree digest so file changes void passing verdicts, and a hard deny on false completion at update_goal via tools/pre-execute with a false-complete counter that blocks after two catches; scope-aware task-level claim handling (strict or goal-only) keeps multi-task Agent Teams loops from deadlocking.

Correct-encoding command runner and mojibake recovery: encoding_run spawns a child process and decodes its raw output bytes (UTF-8/UTF-16LE/GBK auto-detected or forced), and encoding_decode recovers garbled text back to correct Unicode.

Ponytail lazy-senior-dev mode as a native DeepSeek Harness skill, system-prompt section and six slash commands. Always available, level-switchable, no external process.

Checks each step of a plan before you approve it, using command rules plus Jev or an OpenAI-compatible endpoint with logprobs, and marks steps that can't be undone, affect things outside the workspace, weren't asked for or are too vague on the plan review card.

Claude Code-style prompt history, double-Esc picker, selection quote/copy & right-click paste.

Git workflow for DeepSeek Harness: confirmed Web Git actions, slash commands and read-only tools

Read-only work report rebuilt from local session logs: tools used and how often, commands run, files read and written, whether tests passed, failures and their causes, token usage, and whether the task finished.

Readable transcript export for DeepSeek Harness: turn any Session into a Markdown / HTML / ZIP transcript with reasoning blocks, tool calls and results, images and usage — from the Session header or the /export-md command.

Drives a Chromium browser you are already running — Chrome, Edge, Brave, Vivaldi, Opera, Chromium, Arc or Thorium — over one long-lived CDP connection, so the remote-debugging approval is asked once instead of per command. Commands for tabs, navigation, JS evaluation, DOM snapshots, clicks, typing, screenshots and CSS/DOM/pixel assertions go over a loopback HTTP route that the agent calls with the shell tool it already has, so no new tools are registered, and the plugin registers its own usage skill with no runtime dependencies.

Replaces the web profile's pwsh shell executor with a subclass that rewrites every command through `rtk rewrite` before execution for token savings, running the original command unchanged on any rewrite failure (no equivalent, missing rtk, timeout) and keeping sandbox semantics intact.

Declares the OpenCode Go and Command Code route wire protocol and endpoint (so the API address never has to be typed), adds the DeepSeek V4.1 models to the OpenCode Go route and puts them first, completes the Command Code model list from its public catalog, and attaches a per-conversation x-opencode-session header to OpenCode requests (fixes 400 MissingSessionID).