Plugins

Git Changes tab for the DSH web GUI: staging, diffing, AI-drafted commit messages, branch/merge/stash/worktree operations and sync, with live updates via fs.watch plus a changeToken poll.

Maintains reusable named subagent profiles for DeepSeek Harness and runs them with per-profile model, reasoning effort, and agent-preset settings.

Task Runner mode for DeepSeek Harness: the main agent decomposes large tasks into independent subtasks, dispatches them to subagents (default local model) in fresh short contexts, and synthesizes results; concurrency and worker/orchestrator models are configurable in the settings panel.

Jimu (积木): domain-based orchestration plugin for DeepSeek Harness — domain_plan confirmation, per-domain sub-agents with hand-off, and automatic LLM audit.

✨ One-click prompt polisher for the DSH composer — an agent refines your rough draft in a side chat over multiple turns, and you copy the result back when you're happy.

Turns a product or UI request into an implementation-ready design spec with /ui-workflow and /ui-rerun. Analysts run through a read-only ui_analyst tool, and the lead agent writes docs/ui-workflow reports.

Intercepts turn-stopping to force one critique-and-improve round after each completed turn before the agent stops.

Multi-agent isolated debate for DeepSeek Harness: debaters of different temperaments research independently (no herding), then cross-examine, then judges converge to a conclusion. Runs as a background job with per-agent progress, token accounting, wave-parallel execution, and session-referenced resume.

Claude Code style /init command and tool that writes a minimal CLAUDE.md and symlinks AGENTS.md to it.

Captures every plan the agent presents through exit_plan_mode as markdown under <workspace>/.dsh/plans, opens a review window, and keeps a Plans history tab recording the outcome of each plan.

Preset workflow template library: 12 reusable orchestration script templates with wf_template_list / wf_template_search / wf_template_run / wf_template_validate tools and static template validation.

Input redundancy pruning for DeepSeek Harness — drop repeated/redundant tokens before they reach the model (layer ②).

Context compaction / headroom for DeepSeek Harness — compress the live context so the prompt stays small without losing the state that matters (layer 3).

Session context handoff for DeepSeek Harness — stop re-reading and re-deriving context every session. A single skill (layer ①) that keeps the next session warm.

Google's Agent Skills (google/skills) for DeepSeek Harness: 128 bundled SKILL.md skills for Google Cloud, BigQuery, GKE, Gemini, Firebase, ads and analytics.

Resolves unique kebab-case skill substrings in slash tokens, so /oneshot can load game-package-oneshot.

Serves the obra/superpowers skill catalog to the harness and injects its using-superpowers bootstrap as a persistent, compaction-safe system-prompt section.

Skill provider that installs the whole @dennisrongo/skills catalog as one npm dependency, so skills install and update without cloning or linking directories.

Voice control for the DSH web composer: push-to-talk speech-to-text into the input box (with optional auto-send), spoken playback of assistant replies via the Web Speech API, right-click settings popover, and a global Ctrl+M hotkey.

Vision bridge for text-only DeepSeek models: transcribes attached images with deepseek-v4-flash-vision-exp before they reach deepseek-v4-pro, with no third-party dependencies.

Apple on-device Vision tools for text-only dsh models: local OCR (zh-Hans + 30 langs), image classification, face detection, document layout, and a combined describe — 100% offline, no API key, tall-screenshot slicing.

Switchable web-search provider for DSH: native DeepSeek web search or an OpenAI-compatible custom backend. Installation automatically routes DSH web searches to the plugin; only provider, API key and model are required. Published on npm as @lp181818/web-search-advanced@0.1.3.

Resident-index grep for DeepSeek Harness that shadows the built-in grep/glob tools under the same names: repeated content searches hit an in-memory index ranked by access/modification frecency, and glob serves the built-in ripgrep semantics when available, falling back to the index otherwise.

Algorithm learning toolkit syncing Codeforces and AtCoder submissions, profiling weaknesses, generating AI training plans, and tracking spaced-repetition reviews across a 114-lesson algorithm curriculum.