A full star ranking of unique repositories. Stars indicate interest, not quality, maintenance commitments or security.
Project memory for DeepSeek Harness: project files are indexed into searchable summaries as they are read, so after context loss the agent retrieves hits on demand instead of re-reading files, with source citations and BM25 ranking.
把你的任何想法，变成每个平台原生的宣发内容 | Turn any idea into platform-native publicity
Bottom-right quota capsule that auto-discovers every configured provider (DeepSeek, OpenRouter, SiliconFlow, GLM, one-api/new-api, coding plans) and shows balance or rolling 5-hour and weekly usage, with per-provider visibility, alert thresholds and proxy support.
Runtime requirement drift guard that keeps long-running DSH agents aligned with the user's approved goal, constraints, and decisions.
Reproducible Python and R work on DeepSeek Harness, built as plugins.
Reproducible science workbench: agent-driven cells, inline figures with feedback/rerun, manifest provenance, environment snapshots, and publication-grade figure skills (9 bio_* tools + workbench UI).
Fetches any http/https page and returns title, description, main text, heading outline and links, with stats (char/word counts, reading time, language) and top keywords; can download content images to the session workspace for vision analysis and dispatch them to registered analyzers via the scrape.imageAnalyzer service.
Real session deletion with trash and restore, archived cleanup, and batch delete from settings.
Frame-level scan diagnostics for session files (torn/corrupt/empty detection).
DSH（DeepSeek Harness）Skills 管理 + SkillHub 商城插件：侧边栏面板管理本地 skills（用户级/工作区项目级/内置只读），搜索并一键安装 SkillHub 公开技能，附模型工具 dsh_skillhub_search。
Skin marketplace for DeepSeek Harness — search and install community skins from dsh.a2hmarket.ai right in the settings page
DSH Web GUI right sidebar: VS Code-style file explorer plus source control (git status, stage/unstage/discard, commit, diff, commit graph, sync fetch/pull/push, branch management, remote management, git init, multi-repo switch, color-coded file status (untracked/modified/ignored), file-type icons, image preview) with an editable full-file diff view, a syntax-highlighted editor (15 languages), file operations (new/move/copy/delete), and save.
Routes delegated subagent work to configured models: a subagent_model tool whose model enum is the configured aliases, per-route tags and routing guidance injected into the tool schema and system prompt, a wait-for-subagents join tool returning terminal results from outstanding background children, active-model chips in the subagent header and catalog rows, and a Web settings page, namespace-scoped configuration tool, and guided setup skill.
Background task status bar: progress plus live output tail on the chat page.
Tavily search with multiple API keys, key rotation and failover, a usage gauge, and a settings card.
Mineradio glassmorphism theme for the DSH web UI: champagne-gold glass panes, a fluid or wallpaper backdrop, and a one-switch toggle that restores the stock look exactly.
在对话页面直接查看消耗费用（嵌入官方底部状态条，点击看明细）；设置中汇总用量与费用，并可自定义其他模型单价
Layered token-optimization pipeline for DeepSeek Harness: output ladder, MCP lazy loading,compaction driver, cache-hit reporting. Built on real DSH plugin APIs; ~40-60% input saved in long sessions.
A DSH calculator: a zero-dependency, zero-process, pure-function math expression evaluator (sqrt/sin/cos/log/pow…) — so arithmetic stops spawning a bash process every time (especially costly on Windows).
Font engine for the Web GUI: system font enumeration, global and per-component font-size tuning with a Spy++-style picker, settings page.
Usage and cost statistics with peak/off-peak pricing after the 2026-08-17 rate change, a floating summary panel, per-session breakdowns, and a day/week/month/year/all usage heatmap.
DeepSeek usage dashboard for DSH Web: days-left estimate from recent average spend, daily/monthly/cumulative costs with period-over-period comparison, per-model cost ranking, cache savings, peak/off-peak pricing impact, and a draggable floating balance widget.
Low-cost video understanding tool: a video_understand tool turns a Bilibili link / BV / local video into an AVIS info layer (ASR + scene structure + object tracks + YOLO labels) and returns summary + Q&A. Question-driven dynamic routing across layers (L0 ASR / L1 object tracks / L2 key-frame VLM), semantic-layer reuse for repeat questions, per-question budget cap. Self-contained engine, no external dependencies.
Journals every AI file write/edit into before/after snapshots for an inline VS Code review flow (accept/reject hunks, per-file undo, no git).
