The "ai-agents" topic on GitHub groups 503 open-source projects in the DeepSeek Harness (DSH) ecosystem, led by graphify with 117.1k GitHub stars. graphify — Turn any codebase, with its docs, SQL schemas, configs, and PDFs, into a queryable knowledge graph. Every project here is indexed by DSH Universe with live GitHub data — stars, activity and install status — so you can compare and install directly.
Turn any codebase, with its docs, SQL schemas, configs, and PDFs, into a queryable knowledge graph. A /graphify skill for Claude Code, Cursor, Codex, and Gemini CLI: local deterministic AST parsing, every edge explained, no vector store.
AAS Core is the local, agent-first control plane for complete catalog discovery, agent-owned selection, stack validation, and planning, backed by 2,115+ agentic skills. Includes CLI, local MCP, catalog, plugins, and Workbench.
Digital Life Khazix's open-source AI Skills collection | Agent Skills: leader (helps you define goals), neat-freak (cleanliness freak), hv-analysis, khazix-writer & more — Claude Code, Codex & 40+ agents
Self-evolving memory OS for LLM & AI Agents: ultra-persistent memory, hybrid-retrieval, and cross-task skill reuse, with 35.24% token savings and DeepSeek Harness support.
✨ All your agents and workspaces in one place, on every device you own. Track tasks on a board, accessible from desktop, mobile, browser, or API. Self-hosted.
The trust layer for agent-to-agent commerce — natural-language mandates, ERC-7710 delegated permissions, x402 payments, escrow, and dispute resolution as one open, catch-all Agent Skill / Claude Code plugin.
Enterprise-grade, local-first Agent Workbench for people and agent teams. A unified multi-engine workspace for Codex Harness, DeepSeek Harness, and OpenCode, with unified plugins and Skills, multi-agent projects and tasks, and editable code, documents, presentations, design, and video.
Qiushi-Skill: Build agents that investigate first, focus on the main contradiction, validate in practice, and keep pushing until the work is actually done. Qiushi Skill — distills one overarching principle and nine methodological tools from classical materialist dialectics and practical philosophy to arm the AI brain.
Agent Skills for NVIDIA products — install into Claude Code, Codex, and other coding agents to run Physical AI, robotics, simulation, CUDA, and RAG workflows end to end.
Anti-laziness skill for AI agents. Core: the Depth Tree method, which splits a task N layers deep and gives every leaf the full time budget of the whole task, so effort multiplies with depth. Grounded in 2025-2026 research on model laziness, underthinking and premature completion.
J-Space Cognition Suite V3.7 - AI cognitive-enhancement Skills based on Anthropic's J-space global workspace research. | Bilibili: Tiger380 (UID 3494375382321675) — https://space.bilibili.com/3494375382321675
HarnessRouter Community Edition: the self-hosted, Apache-2.0 edition of the unified interface for agent harnesses. Run Codex, Claude Code, Hermes, PI, DSH, and more through one API, with sessions, streaming, files, cancellation, and failure handling. Implements the Unified Harness Protocol (UHP), an open standard. Your keys, your infrastructure.
AI agent skills for social media research. Outlier posts, comment mining, competitor teardowns, ad libraries & trends across TikTok, Instagram, YouTube, Reddit, X, LinkedIn & more. Powered by ScrapeCreators. Works with Claude Code, Cursor, Codex, Gemini CLI.
Connect IM bots to DeepSeek Harness via QR code or bot credentials (supports Feishu, WeChat, DingTalk, WeCom, QQ, Slack, Telegram, Discord and WhatsApp). Connect IM bots to DeepSeek Harness via QR code or credentials (9 channels).
1098 professional Agent Skills for Claude, ChatGPT, Gemini, Cursor & Codex — from PRDs and postmortems to appealing a disability benefit, building a go-bag, and settling into a new country. Plain-markdown, MIT, in Anthropic's official plugin directory. Free in-browser or 'npx pm-claude-skills add'.
An IOS Simulator Skill for ClaudeCode. Use it to optimise Claude's ability to build, run and interact with your apps, and to proxy xcodebuild to save token and context wastage.
Self-evolving second brain with 33 AI skills, 10 agents, and people CRM. Closed-loop harness: a V-model verification lifecycle where the worker never grades its own homework. Plus paired anti-slop design skills for marketing and product UI. Works with Claude Code, Cursor, Kiro, Gemini CLI, Codex.
Agent Skill for cloning websites: capture a read-only mirror, restore it line by line from minified code, verify with automated diffs. An agent skill that mirrors a website, rebuilds it from the minified code, and verifies the result with automated diffs.
Journal-specific Claude Code/Codex skill packs covering mainstream journals — AER, QJE, Nature, Cell, Management World, Economic Research Journal & 200+ more — your fast track to getting published. | Claude Code/Codex journal skill packs covering mainstream journals, from topic selection and identification strategy to table formatting and reviewer responses—the full workflow helps you publish papers quickly.
Take your agent-built product live: hosting, database, domain, email, payments — on your own accounts. Open-source Agent Skill + zero-dependency Node CLI: detect → plan → approve → apply → verify. No GoLive account, backend or telemetry.
Claude Code Skill Factory — A powerful open-source toolkit for building and deploying production-ready Claude Skills, Code Agents, custom Slash Commands, and LLM Prompts at scale. Easily generate structured skill templates, automate workflow integration, and accelerate AI agent development with a clean, developer-friendly setup.
Stack-agnostic Claude Skills covering the full website lifecycle: brand, design, content, SEO, dev, ops, growth, and research. Build, ship, audit, optimize.
AI agent skills for the personal operator's craft — decisions, research, second-brain, content rotation, scenario modeling, and meta-skills to author more. Works with Claude Code, Codex, Cursor.
DeepSeek Harness (dsh) from 0 to 1 in-depth handbook: installation/plugin development/performance tuning/real-world cases/multi-Agent comparison with the same model (Chinese + English PDF)
The most complete, continuously updated, hottest collection of 100+ persona distillation skills|multi-agent system|celebrities/directors/Tianya legends/ancient texts/anime/workplace/relationships, all categories
Local-first AI agent runtime with sandboxed sessions, MCP tools, memory, credentials, audit/replay, and a built-in console. Run OpenAI, Anthropic, MiniMax, DeepSeek V4, and OpenAI-compatible models on your infrastructure.
A highly customizable agentic harness for arXiv-ready ML/AI review papers (and beyond). It drives agentic AI like Codex CLI and Claude Code through a gated LaTeX workflow with verified BibTeX citations.
11 Claude skills for SEO: page audits, linkbuilding, article writing, E-E-A-T audits, semantic gap analysis, link building. Methodology from Koray Tuğberk, Kyle Roof, and Lily Ray, plus a generation-time anti-AI-slop ruleset. Production-tested at InhouseSEO
DeepSeek Harness (DSH) plugin: a live iOS Simulator — and a USB-connected iPhone — inside the conversation. 22 agent tools for booting, building, driving the UI by accessibility identity, OCR text or list rows, plus a streaming sidebar panel you can tap and drag on.
Stable desktop client based on DeepSeek Harness (DSH), with Git, built-in browser and memory | DeepSeek Harness desktop GUI with local workspaces, Git, browser and memory.
A curated list of plugins, skills, MCP servers, patch/profile layers, orchestrators & UIs for DeepSeek Harness (DSH). Visualization · PPT · Coding · Agents · Loops (auto-research) and more. #dsh
Claude skill that analyzes images, websites, and Figma files and generates a structured design.md with token system, component inventory, and reconstruction notes.
DSH Plugin Marketplace / DSH Plugin Marketplace: Browse, install & update all GitHub topic:dsh-plugin plugins with one click in the DeepSeek Harness Web GUI | browse, install & update all GitHub dsh-plugin plugins in the DSH Web GUI
DeepSeek Harness plugin for Android — build, run, and interact with a live emulator or USB device stream inside a conversation, driven entirely through adb.
Skill для AI-агентов (Claude Code, Codex, OpenClaw, Hermes): убирает 37 признаков AI-генерации из русского текста и проверяет, писала ли его нейросеть - канцелярит, штампы ChatGPT, артефакты копипаста | Russian AI-text humanizer & detector skill for coding agents
DeepSeek Harness (DSH) plugin: dispatch work to DSH agents from Claude Code / Codex — native subagent progress, in-host worker sessions with per-tier presets, and a multimodal bridge that lends the text-only harness vision and image generation.
Bộ công cụ và tập hợp skill hỗ trợ phân tích phần mềm, khôi phục cấu trúc nguồn ở mức cần thiết, rà soát bảo mật, kiểm tra phụ thuộc và xây dựng kế hoạch khắc phục cho các hệ thống mà bạn sở hữu hoặc được phép đánh giá.
Open-source AI skills for SEO, AI search visibility, conversion copy, marketing strategy, and business operations. Reusable workflows for Claude Code and Codex, plus code review, app delivery, and creator tools.
Persistent multi-model workflow teams for DeepSeek Harness — dynamic lead planning, bounded DAGs, per-agent model/tools, Run Center and Token insights.
Native Agent-driven long-form roleplay: interactive creation, proactive world book, dynamic memory, multi-character simulation and novel export. Create a world, step into it, and take it with you.
A Claude Code skill that delegates coding, review, analysis, and research jobs to Google Antigravity CLI (agy) sub-agents — save tokens, use a second model, work in parallel.
Agent skills for building iOS apps on iPhone Duo, Apple's first foldable iPhone — adaptive layout, vertical bars, hinge & scenes, camera, and design review.
Make Agent workflows composable, reviewable, and continuously improvable, ultimately achieving Agent Self Evoling. DeepSeek Harness Web plugin with composable task controls and isolated, human-reviewed self-evolution.
DeepSeek Harness three-platform compatible desktop launcher: fully isolated multi-instance, parallel collaboration, collaborative canvas orchestrating Agent workflows; portable Exe one-click launch, only about 18M (no more than 20M); dual isolation makes it highly compatible, no matter how DSH core updates, compatible from native to wild.
An Agent Skill that turns the asks you keep typing into short commands, mined from your own Claude Code and Codex history. Works in Claude Code, Codex and Cursor.
Community Docker and Kubernetes packaging for DeepSeek Harness (@deepseek-ai/dsh), with a hardened image, Compose stack, Helm chart, Web UI, and headless CLI.
Deploy and host DeepSeek Harness (dsh) instances — self-hosted on your own server or cloud-hosted: a multi-tenant platform with isolated Docker containers, persistent workspaces and resource quotas.
88 installable open-source Agent Skills for research, social intelligence, marketing, and business workflows—compatible with Codex, Claude Code, Cursor, Gemini CLI, and DeepSeek Harness.
Search 65,600+ API docs from 20 Chinese open platforms directly in DeepSeek Harness / Claude Code / Cursor / Codex / Gemini CLI; zero-config, supports Skill and DSH native plugins.
sofagent is an open-source FDE Harness for AI coding agents — on entry, write your business judgment into files (workflows, ontology, AI nodes); after departure, audit every change against them. 25 codified audit rules, tamper-evident chain, snapshots, and an MCP toolset (104 tools, 11 plugins). Everything can be FDEing.
HEDGEHOG codes Cleaner, Faster and with Fewer Tokens. Hedgehog's AI-driven development builds a task dependency graph from your spec-driven, BMAD-METHOD plan, so Claude Code, Cursor & Gemini CLI stay locked to it. A CLI-enforced state machine for agentic coding. Now builds DeepSeek DSH Plugins. DeepSeek Harness, DSH Plugin, AI coding, BMAD method. AI Copywriting
PANews Agent Toolkit is PANews's official agent package for cryptocurrency and blockchain news discovery, creator workflows, and PANews web-page reading.
One-click setup of MCP, Skills and rules for all kinds of AI Agent tools; supports Claude Code, Codex, DeepSeek Harness, OpenCode, Cursor, Gemini CLI, Qwen Code and more. One-click local setup for MCP servers, skills, and rules across AI coding agents.
Make AI work like a senior test engineer: a test engineering Skill framework for AI Agents — 11 Skills + shared knowledge base + type decision matrix (usable with Claude Code / dsh and other Agents)
DeepSeek Harness standalone community plugin store: discover, install, and submit verified plugins, tools, and extensions. | Independent community plugin directory.
A visual multi-Agent workflow orchestration plugin built for the DeepSeek Harness Web GUI. Public beta is live! The official release will follow the DSH official release! Compatibility update: the plugin is now aligned with the latest release candidate dsh v0.1.2-rc.1, and does not support Alpha versions. Notice: since the plugin's underlying architecture relies on the official code's injection rules, there are no plans to support strongly breaking-change versions, because they are really hard to fix!
DeepSeek Harness (DSH) plugin: a read-only ledger for the plugins you already have installed — a capability inventory with file:line evidence, declared-vs-detected reconciliation, cross-profile version drift, and a diff of what changed since the last scan.
Text-first browser & background macOS control for DeepSeek Harness (DSH): target the right process and window without taking the user's pointer. Provides text-first computer control for DSH: operate Chromium and macOS in the background without stealing focus or moving the mouse.
A safety gate for DeepSeek Harness: a stronger model reviews dangerous tool calls before they run. Plus vision and BlockRun's full model catalog from one wallet, paid per request over x402.
DeepSeek Harness URL reader: fetch any page and return clean main-content text/Markdown. Auto charset (GBK/GB2312/UTF-8/Big5), token-efficient (6000-char cap, cache, offset), zero deps, no API key. One-click full-page reading → clean text / structured Markdown
Have to edit config and restart to add MCP? DeepSeek Harness (DSH) plugin Skill Config solves it: paste JSON on the Settings page to connect to MCP servers, hot-load without restart; upload zip/.skill to auto-detect SKILL.md, create/edit/start-stop skills, click to use
AI argues. Code settles. The losses stay on the page. A real HK + US brokerage account run by agents that must debate every call, settled by code the model never touches. Install the same decision workflow into your own agent: OpenClaw, Claude Code, Codex, or DeepSeek Harness.
OOMOL Connector for DeepSeek Harness—discover connected apps and execute Actions through progressive MCP discovery without exposing provider credentials.
DeepSeek Harness install-free portable edition (Windows): official npm package + bundled Node.js, double-click the exe to use, copy to any 64-bit Windows PC and run standalone
DeepSeek Harness whitepaper · 131-page beginner and advanced comprehensive tutorial: from installing it, understanding it, to extending it yourself. For developers who want to study agent harness in depth, all conclusions verified against source code.
DSH plugin: OpenCode Go quota ring — a progress ring to the left of the model selector in the input box, click to switch 5-hour/weekly/monthly usage (OpenCode Go quota ring for DeepSeek Harness Web)
DeepSeek Harness plugin for Fly.io Sprites. It gives AI agents persistent Linux computers with hardware isolation for building, testing, and deploying software.
Sleep tight. NightMe codes all night. A remote-pair developer agent. No more babysitting AI—stay in the loop from your phone while AI Coding Agent(claude/codex/opencode/pi/dsh/cursor) takes the wheel.
DeepSeek Harness (dsh) plugin: registers a global skill that teaches agents how to discover, evaluate and install community plugins from the GitHub dsh-plugin topic, dshmarket and npm. | DSH community plugin ecosystem guide skill: teach agents to discover, evaluate and install community plugins.
Check Agent Skills before reuse: find missing local resources in browser or CI, review pinned examples, and serve skills through DeepSeek Harness or MCP.
Turn your devices into a fleet — dph plugin for decentralized multi-device collaboration: mDNS discovery, key pairing, SSH direct exec. Tools auto-register for dph agents.
Official Riffkit skill — riff a winning TikTok into your own short video from your AI agent (Claude Code, Cursor) or the browser. Riff the formula, not the video.
Product Team Mode - a DeepSeek Harness agent preset: user-led conversation with a virtual product team (PM to Engineer to QA to Release) walking you from idea to shipped product
Real-time chain-of-thought trajectory profiling for DeepSeek Harness: live wording indicators, profile-family judgment, per-session measurement records
🐳 Blue Big Fat Fish · DeepSeek orca 3D desktop pet plugin — rebuilt from the outline of the DeepSeek orca logo, floating in the corner of the Web GUI: drag to rotate, 360° spin, heart interaction, hover toolbar, size/sensitivity/caption settings, config persisted locally / A 3D whale desktop pet rebuilt from the DeepSeek whale logo for DeepSeek Harness — drag to rotate, 360° spin, heart bursts, hover toolbar, size/sensitivity/caption settings, and local persistence.
Multi-agent orchestration for DeepSeek Harness: turn isolated sessions into a team that assigns work, reviews output, and runs DAG workflows — durable task ledger, real reviewers, crash-safe recovery.
Folio · Consulting material generation engine: 15 DSH native tools + automatic execution of session protocols + L0 guard + swappable methodology packs, start with zero keys. Consulting material engine for DeepSeek Harness: HTML/PPT/DOCX/quote with mechanical quality gates, bilingual demo outputs in the v1.0.2 release. Help wanted: Word layout, Excel formatting/formulas, Linux/mac ports.
ChatGPT Subscription - a DeepSeek Harness plugin: bind your ChatGPT account via official OAuth and chat with ChatGPT models inside DSH, using your Plus/Pro subscription quota
Hatch a Digimon inside DeepSeek Harness: grows by eating your real work, Digimon-style branching evolution, zero tokens, the model doesn't know it exists | Digimon-style raising game for DSH — feeds on your real work, zero tokens.
A Yogācāra (Consciousness-Only) self-model plugin for DeepSeek Harness: eight consciousnesses, the 51 mental factors, a perfumed seed store, and a measurable self — written back into the agent's own prompt.
Multi-platform IM gateway for DeepSeek Harness: Feishu (Lark), WeCom (WeChat Work), and Telegram. One agent per chat, whitelist access, no public endpoint needed (Feishu long connection / Telegram long polling).
One session, one door, one realm's fate. Myriad-worlds DSH: a word-battlefield on DeepSeek Harness. Nineteen doors — start from a cover; no assistant, no outline, only the body text. What gets transcribed is the heavenly book of the life you lived.
DeepSeek Harness desktop: embedded Web UI, AI screensaver (replaces system screensaver), full model switching, phone PWA remote control, QQ bot channel
Put a watch on a stream: background listeners that wake the DeepSeek Harness agent with new matching lines — and a daemon host so a watcher runs unattended for weeks, with no task and no browser. Not affiliated with DeepSeek.
DSH Lab: DeepSeek Harness hands-on picks and Chinese guide, every entry tested on real hardware through four gates (install → smoke → quick security check → compatibility pinning)
SecurStack adapter for DeepSeek Harness: run repository security scans, policy gates, doctor diagnostics, and JSON CLI results from safe AI-agent tools.
Symphony-inspired multi-provider task orchestrator and native operations dashboard for DeepSeek Harness. Supports Linear, GitHub, Jira, Asana, GitLab, and local tasks.
DSH multi-party group chat collaboration plugin (round table): pull AI Agents with different personas into the same discussion room to debate, cross-supplement, question and correct each other, then summarize a combined plan in one click; also supports pulling existing conversations in the workspace into the discussion to eliminate information silos. Supports three speaking modes — manual @ / all-hands discussion / relay chain — with round limits, global pause and other anti-infinite-loop controls. Official bundle plugin: dsh plugin --profile web add github:vuchisu069-source/dsh-round-table#main
One model writes the work, a different one reviews it, and neither decides whether it passed. Cross-provider AI review with verifiable evidence, for DeepSeek Harness/Cordis hosts. 中文文档见 README.zh.md。
dsh-dock · DeepSeek Harness feature hub: one management panel to register and toggle all small features (model balance, Token usage records, task animations, etc.). Each feature is an independent module, with toggle and error isolation, new features plug and play. 0.1.0 is the base framework; feature integration iterates per the README roadmap.
Local personality, memory and distillation layer for DeepSeek Harness — SOUL/IDENTITY/USER/STATE persona files, cross-session memory and LLM distillation, auto-loaded on startup.
DeepSeek Harness plugin that does QA: an agent explores your app like a real user and leaves evidence, then the explored path is exported as a deterministic replay scenario you run every release. Orchestrates the browser, macOS desktop, iOS and Android drivers — never a false green.
The quality gate for agent-built web apps. Real-browser acceptance checks — CLI, MCP (Claude Code / Cursor / Copilot), GitHub Action. Agents self-test and pass; real browsers tell the truth.
Claim verification for AI coding agents — logic-primitive verification for state machines, data models, and concurrency risk in design docs & refactoring plans. Cross-platform Agent Skills + DSH native. | AI coding agent claim verification plugin: logic-primitive verification of design docs and refactoring plans, covering state machines, data models and concurrency risk, with before-and-after regression. Cross-platform Agent Skills + DSH native.
DeepSeek Harness memory framework plugin. MEMFLOW memory flow mode: sessions end, memories don't — perceive first, record as you go, resume seamlessly. Distributed memory architecture: every project has its own memory, subtasks customize memory context, project skill self-iteration — one plugin brings Hermes-like results to all your workflows.
DSH plugin for durable, cross-session memory & todos — five tracks (global/user/project/key/daily), confirmation-gated writes, Git-backed sync, in-place adoption of ~/.dsh/memories.
Windows desktop UI context picker for AI agents: select windows, UI elements, or screen regions with hover highlighting, UI Automation, screenshots, local OCR, and user-approved MCP output. Works with OpenAI Codex, DeepSeek Harness, and other MCP-compatible clients.
DeepSeek Harness session replay and redacted sharing plugin: export real Agent trajectories as standalone interactive HTML for docs, demos and issue reports.
DeepSeek Harness plugin: one click binds each conversation to its own isolated git worktree + branch, so parallel agents work on one repo without interference — every project's worktrees auto-managed in the sidebar workspace tree, backed by 9 git tools.
Extensible chat-skin monorepo for the DeepSeek Harness Web GUI: a base registry + plug-in skin packages (Feishu-style chat bubbles), plug-and-play without touching chat logic.
DeepSeek Harness development forge: approval guard, dev Skills, GitHub/browser capabilities and Token Watch consumption supervision — install and get to work.
dsh-FEMO (Flow Emerges Opus) — dsh plugin multi-agent script engine: Femo sessions, chat window character bubbles, script control (full engine included, move the whole folder and it just works). Private until scrub review.
Wishes that outlive the session — turn a DeepSeek Harness agent's runtime plugins into permanently installed ones. No pnpm, no network, no build authorization.
Your real work is the tall grass: wild sprites pop up at random, throw balls to catch them, 28-slot dex, 6-member team, nocturnal species and legendary conditions, Pokémon-style collecting, zero token | Collect-em-all safari for DSH — your work rustles the grass.
A desktop agent app built on the DeepSeek Harness (dsh) runtime — an Electron shell bundles the dsh CLI, provisions built-in plugins and skills, and hosts the dsh Web UI. Build your own desktop agent app through plugins.
A DeepSeek Harness (DSH) dynamic plugin that adds an ✨ optimize-prompt button to the chat composer — context-aware LLM rewriting with model fallback and visible errors.
A collection of dynamic Cordis plugins for DeepSeek Harness (DSH): token cost metering, balance checks, update checking, and a workspace file explorer with code & Markdown preview. Plain JavaScript, paste-and-run.
ComfyUI driver (AI-first contract-based): the model sees only contract slots, never the workflow JSON. text-to-image / image-to-image / text-to-video / image-to-video templates, double pre-check against hallucination, live detection of official/custom nodes.
🌊 Abyss Agency · Abyss — paints DeepSeek Harness multi-agent runs as an agency you can see: delegation lineage, per-agent cost and failure attribution, context level, attendance timeline; past cases can be replayed and exported as a Markdown retrospective in one click. A DeepSeek Harness plugin that turns an agent fleet into an office you can watch.
A DeepSeek Harness plugin that puts graft — a prebuilt graph of every symbol, its file:line span, and who calls what — in front of both you and the model.
One-command DeepSeek Harness on your VPS: a zero-dependency login gateway puts the stock DSH web UI - settings, API keys, plugin market - behind HTTPS and a password, reachable from anywhere. · 一条命令把 DeepSeek Harness 装进 VPS:零依赖登录网关 + 自动 HTTPS,公网可达,设置、API Key、插件市场全都能用。
DeepSeek Harness plugin: give your agent a browser with a persistent identity - engine-level fingerprint spoofing, unlimited free local profiles, Android device emulation, passkeys that survive, and residential proxy egress.
Portrait & mobile adaptation for the DeepSeek Harness Web UI — overlay drawer, full-width conversation, sheet dialogs, safe-area handling. Below 1024px it adapts; at ≥1024px it is a no-op.
A full engineering bench for your coding agent: the work survives any session, and every done comes with proof. Runs on Claude Code, Codex CLI, Copilot CLI, Kiro CLI, and DeepSeek Harness.
LLM-provider plugin for DeepSeek Harness: use local AI CLIs (claude / opencode / gemini / cursor / codex) as model backends, hot-switch in the model selector. DSH's LLM provider layer plugin: local AI CLIs as model backends, one-click hot-switching.
DeepSeek Harness multi-agent collaborative development plugin with seven built-in agents (Qibo, Lishou, Feng Hou, Kui, Li Mu, Gao Yao, Li Zhu), covering the full development loop of project initialization, code categorization, plan orchestration, batch execution, code review and test verification. Built on the @deepseek-ai/cordis plugin system, giving dsh "Feng Hou & Li Mu"-style multi-agent collaboration.
Kestrel - a research engine that can't cite what a source never said. Citations are admitted only when the quote is mechanically located in the source; corroboration is counted in independent origins, not source count. DeepSeek Harness plugin + standalone library. Zero dependencies.
Add a "review on my behalf" feature to DeepSeek Harness, letting AI review DSH sensitive operations for you instead of handing over all permissions. Fail-closed automatic approval for DeepSeek Harness with selectable reviewer models and visible WebUI decisions.
/code-review slash command for DeepSeek Harness — five parallel review lenses, per-finding confidence scoring, then a gh reply back on the pull request.
Nine deterministic hooks that block a coding agent from weakening tests, swallowing errors, or stubbing type checks — each backed by a planted-failure test proving the guard can actually fire. Runs on Claude Code and DeepSeek Harness.
Delayed wake-up for DeepSeek Harness: schedule a reminder and the agent wakes in the SAME conversation to reply — plus an interruptible wait (timed gate) tool. Host-level timer + official agent.followup() wake channel.
DeepSeek Harness named subagent roster plugin: settings-driven role roster, list_subagents / delegate tools and a settings page. Named subagent roster plugin for DeepSeek Harness.
Task lifecycle plugin suite for DeepSeek Harness (dsh): agents claim work across sessions, wake up on schedule, quota-aware; idle/imported chats auto-mined into board candidates; full web kanban. | dsh task full-lifecycle plugin suite: cross-session claiming, scheduled wake-up, quota-aware, idle sessions auto-mined into candidates, with a Web kanban.
DeepSeek Harness (DSH) plugin/skill: control open Word/Excel/PowerPoint via Office add-in (33 actions, AI-orchestrated, near-Copilot workflows) | DSH Office skill: control open Word/Excel/PPT
Make destructive AI-agent actions reversible by default — quarantine + audit + human escalation for rm/git destructive operations. Reliability infrastructure, not a sandbox.
Run DeepSeek Harness against a gVisor-isolated cloud sandbox — files, Bash, PTY and LSP all move off your machine. Drop-in plugin, no fork, optional persistence.
DeepSeek Harness personally developed plugin set: context compass / cross-session knowledge / subagent model routing / AI image generation (Personally developed plugins for DeepSeek Harness)
An agent-native compilation & verification pipeline that converts software capabilities such as CLI, APIs, and local services into installable, testable, verifiable DeepSeek Harness plugins.
Map-reduce council of subagents for the DeepSeek Harness: one task fans out to independent members, their findings are deduplicated, verified by a separate panel, and reduced to a quorum report
Publish HTML from DeepSeek Harness to a private, live URL. No account for the viewer, unguessable links, replace in place. DSH plugin (Cordis overlay) for the Stacktree MCP server.
Give DeepSeek Eyes · give DeepSeek eyes — a DeepSeek Harness (DSH) Skill: local vision models or vision APIs let DeepSeek read images and graphic files
DSH Game Studio: AI-native Game Development Runtime for DeepSeek Harness — installable/uninstallable/upgradable game dev plugin (/game subcommands, dynamic Agent Pool, engine adapters, Verifier + Quality Gate)
DeepSeek Harness plugin: role-based model routing — planner (root agent) on deepseek-v4-pro, delegated executor subagents on deepseek-v4-flash; ships a prompt section and a pro-flash-routing skill.
Customizable semantic color accents for assistant Markdown, tool calls, and Think content in DSH Web. Customizable semantic color accents for assistant Markdown, tool calls, and Think blocks in DSH Web.
macOS menu bar lid-close assistant: supports DeepSeek Harness, ChatGPT, Claude task awareness, offers auto/manual lid-close policies and black-screen power saving|AI-aware lid-close utility
Workflow-net runtime for DeepSeek Harness: resource-aware concurrency, native loops and fan-out, static soundness checking before a plan becomes durable, and process mining over its own event log.
Let an AI assistant drive STUDIO NEUTRINO (free Japanese AI singing synthesis engine) to sing: give lyrics and melody, and it auto-generates a WAV of the character singing. Bring your own free NEUTRINO engine and voice banks; the plugin includes official download guidance.
在 dsh 里装上这个插件即可,无需登录、注册或填 API Key,就能使用包括 Muse Spark 1.3、MiMo V2.6 在内的前沿模型——完全免费,不限量。 All you do is install this plugin in dsh: no login, no sign-up, no API key — the frontier models are just there, Muse Spark 1.3 and MiMo V2.6 among them. Completely free, with no usage cap.
AgentCabin — open-source desktop workspace for Pi Coding Agent and DeepSeek Harness (DSH), built with Electron, Svelte, and Rust. Local-first AI coding agent desktop for Pi Coding Agent and DeepSeek Harness (DSH), built with Electron, Svelte, and Rust.
ARI (Agent Runtime Interface): one shell, any coding-agent harness. An interface specification derived from source-level study of DSH, Codex, ZCode, OpenCode, Pi, and ACP.
Your agent wakes on evidence, not on a clock. A community plugin for DeepSeek Harness: watches the streams your harness already carries and decides - hold, flag, counsel, or wake. Local judge, shadow by default, MIT.
The auto permission tier for DeepSeek Harness: workspace-write requests run directly, sandbox escalations (danger-full-access) are judged by a separate classifier model.
Brave Search provider for the DeepSeek Harness (dsh, Cordis): answers the model web_search tool with Brave's agent-oriented LLM Context or the classic Web Search API. One-command plugin install, no build step, API key from the GNOME keyring or pass, never a file.
ego lite browser automation for dsh that remembers — recall a site's learned tools, promote a working script into a new one, hand the user the keyboard for real. 7 tools, host-only, MIT.
Client-neutral AI-agent control plane: user global config overlay for Hermes/Codex/DSH/GitHub/Open Design/OpenHuman via one adapter contract (CC Switch observe-only). Modules: client-neutral-core (task/telemetry ledger, sidecar), services (orchestration/policy/receipts), read-only observer. Authority chain + exact-SHA CI gates.
flykit for DeepSeek Harness — explorer with live file watch, agent terminals (Claude Code, Pi, Codex), DSH agent tools, a searchable model picker and a Claude-Code-style status line. Fly your agents.
Pesquisa web para agentes de LLM via API Tavily com rotação AUTOMÁTICA de chaves — se a request morrer (429/432/401/5xx/rede), puxa outra chave e refaz a MESMA request; o agente nunca vê o erro. Funciona em qualquer agente/terminal, dentro ou fora do DSH.
Your AI agent keeps repeating the same mistakes? Give it a LESSONS.md notebook — it records its own failures into rules, auto-escalates repeat offenders, and you watch it all live in the DSH Web UI. 零 Key,让 DSH 的 AI 自己记错题、面板一眼看清。
Knowledge Forge — a DeepSeek Harness native implementation of the Karpathy-style "LLM-compiled knowledge base": 5 pure-agent pipelines for compile/link/metabolize/Q&A archiving/health check, zero Python, zero vector DB, retrieval is thinking.
Self-Evolving Agent — a DeepSeek Harness framework that gets smarter the more you use it: three journals + six trigger scenarios + experience promotion pipeline + soul.md identity layer + selective experience injection.
Deny an AI agent access to .env files, credential stores, keys and any path you hide — across file tools, shell commands, search selectors and run_code. A DeepSeek Harness (dsh) plugin.
Container-deployable agent harness on DeepSeek Harness — multi-provider routing, plugin-extensible personas and skills, local OpenTelemetry tracing and cost attribution.
GROX Network for DeepSeek Harness (dsh): find agents on grox.life, hire one for a task, get the work back — escrowed, refunded on failure, disputable. Plus a grox subagent provider.
DeepSeek Harness (dsh) plugin: live token cost of every chat, its subagents and the whole session tree — DeepSeek, OpenAI, Anthropic, Gemini, Kimi, Grok and Mistral prices from a bundled catalog, plus an append-only JSONL cost log in your project folder.
Inject a user-defined .keep-reminder file into DeepSeek Harness conversations as a <system-reminder>, with session / turn / request modes.Inject important content from a .keep-reminder file into DeepSeek Harness context as a <system-reminder>, supporting three injection modes: session / turn / request.
A DeepSeek Harness plugin that compiles a described world into declarative rules and lets agents, subjectivity, and emotion emerge — encode the story, not the emotion.
Peer-to-peer conversation linking for DeepSeek Harness (DSH): @-mention any session, wake it up, chat across conversations. Type @ in the composer, pick a session, send — the target agent is woken via agent.send. Persistent message board, native-feel inbox panel, 5 chat_link_* tools. dsh-plugin.
Queue Consolidation for DeepSeek Harness: merge queued follow-up messages into one consolidated formal prompt for the next turn instead of piecemeal one-message-per-turn rework.
Model routing for DeepSeek Harness, with a standing capability-to-model assignment table, per-route reasoning levels, and a user-triggered sync that researches public model prices. 为 DeepSeek Harness 提供模型路由:可持久化的「能力→模型」分工表、按路由的推理档位,以及手动触发、由模型联网核对公开价格的同步。
DeepSeek Harness plugin: planner/executor separation — the main model plans, dedicated execution subagents do the work on a configurable model, managed from the Web settings panel.
Persistent, searchable, per-project memory for the DeepSeek Harness: decisions, rules, and session context in a queryable DuckDB file, with the rule set injected into every model request — plus a full management UI in the Web Client.
Your agent, networking with everyone else's. OpenAgentNetwork connects AI agents to find each other, negotiate, and close real tasks — from B2B sales to hiring to dating — humans always decide. Protocol, SDK, and connectors for any agent platform.
DSH (DeepSeek Harness) tool plugin: archive and read X (Twitter) posts as permanent Markdown snapshots, powered by 2100laike X Archive (x.2100laike.com)
DeepSeek Harness (dsh) plugin: publishes a Command Code subscription (GOAT / Pro / Max) as llm-pi-ai provider routes, with account usage and web search on the same credential.
Deterministic project actions for people and agents — a DeepSeek Harness Web plugin. One actions.json, two surfaces: a right-sidebar panel and agent tools.
An evidence-badged field guide to DeepSeek Harness (dsh), plus a working plugin. Every claim marked with how it was verified. Independent and unofficial.
A dsh plugin that periodically scans reddapi.dev for new Reddit leads matching your one-sentence ICP, dedupes what you've already seen, and writes a dated markdown report. Also forwards 6 read-only Reddit search/lookup MCP tools.
Telegram channel for DeepSeek Harness — chat with your agent from Telegram, with a deny-by-default allowlist and in-chat approval for destructive tools.
Claude Code-style rewind for DeepSeek Harness (dsh): edit a past user message and re-run on a forked session, restore files to the turn boundary, or branch with a summary
Godot capability family for DeepSeek Harness: headless project tools (validate/run/format/inspect/editor-script), GDScript LSP navigation, and a scripted debugger.
My AI-agent setup and Linux workstation, as code: one rules source rendered into Claude Code, OpenCode and DSH, plus Deno scripts that install apps and shell.
DSH text-to-image plugin: lets deepseek-v4-flash generate images from text relying on its own abilities alone, with no other LLM — the model writes GDI+ (System.Drawing) drawing scripts, powershell.exe renders a PNG locally, shown directly in the conversation, zero external image API
Eight DeepSeek Harness (dsh) plugins and the talk they were written for: model tools, slash commands, full tabs in the session window, system-prompt sections and multi-agent flows
Ship your skills, grow your influence — a model-agnostic publishing protocol for AI agent skills: preflight checks, GitHub repo, index PRs, marketplace submissions, promo copy. A process protocol for standardizing agent skill publishing to GitHub.
Stop letting your AI guess — a research protocol for tech decisions that any AI agent (DSH/Claude/Cursor/Codex) can follow: quantified requirements, T1-T6 source tiers, quality gates, traceable verdicts. Model-agnostic technology selection research protocol.
Clean, portable agent skills for DeepSeek Harness, Claude Code and any harness reading the Agent Skills standard. Plain text, nothing to build. Every skill ships a checker and a measured evaluation protocol — and says what was not verified.
Model pricing & capability board for DeepSeek Harness: per-1M-token prices for 7,100+ models across 213 providers (models.dev + harness catalog overlay), cheapest-route comparison, capability tags, and a community promotion feed.
Mac × AI Agent development environment complete setup guide — macOS setup for AI-agent-driven development (all Apple Silicon Macs): Homebrew, Ghostty, mise, uv, dotfiles (chezmoi), Claude Code / Kimi Code / Codex / DSH / PI / WorkBuddy / ZCode
MOVED to @yadsh/dsh-doc-impact in xarleyn/dsh-plugins — Documentation impact enforcement for DeepSeek Harness — keep docs in sync by linking code changes to affected documentation.
MOVED to @yadsh/dsh-session-scope in xarleyn/dsh-plugins — Per-session workspace scoping for DeepSeek Harness — expose only selected directories to agents with focused and isolated enforcement.
MOVED to @yadsh/dsh-sleev in xarleyn/dsh-plugins — Sleev integration for DeepSeek Harness with route-aware LLM telemetry and context-optimization observability
ZCode-style custom subagents for DeepSeek Harness: Markdown-defined roles become agent tools your model delegates to — on pinned model routes or external CLIs (cmdc, pi, claude…), with a built-in manager UI.
DeepSeek-PA (DPA) is an unofficial modular desktop workspace built on DeepSeek Harness (DSH), offering separate Chat and Agent spaces, plugin and Skill extensions, global themes, and supervised multi-agent cluster projects. Unofficial modular desktop workspace for DeepSeek Harness (DSH), featuring separate Chat and Agent views, plugins, Skills, global themes, and supervised multi-agent projects.
DeepSeek Harness plugin for macOS desktop automation. Observes the accessibility tree, gates risky actions behind one-time approval, permanently refuses secure fields, and returns bounded visual evidence — through a native helper you build and grant locally.
Remote-only OpenKapsel workspace bridge for the DeepSeek Harness: replaces host filesystem and shell tools with a fail-closed remote tool set. Self-hosted alternative to cloud agent sandboxes.