Custom agents

Multi-agent browser MCP runtime for isolated, persistent, parallel Playwright sessions. Run multiple AI agents without shared tabs or browser state.

The multi-agent coding loop that covers every angle — one model builds, independent AI models review it from every direction, tests decide, you hold the final gate. Privacy-enforced, anti-drift, self-improving.

Self-hosted Personal Agent

An enterprise workbench for the DeepSeek Harness: AI employees, knowledge bases, skills, MCP servers and DSH plugins, all manageable from a running session.

Make your browser speak STONKANESE - Turn any Chromium based desktop browser Agentic!

A Git GUI plugin for DeepSeek Harness Web UI

GUI for pi - Sessions, Streaming, Tools, and Thinking without leaving the agent stack.

AI-native developer collaboration platform unifying team chat, ticket lifecycle, and AI code execution. Two-stage agent pipeline (Planning → Code) with human approval gates, running entirely on local open-weight LLMs via vLLM. FastAPI/PostgreSQL modular monolith, Redis, Qdrant, React/TypeScript, Docker sandboxing.

Open Vibe Architecture (OVA): architecture-as-markdown framework and Python CLI for coding-agent repositories.

Agentic AI coding tool for your terminal. Reads and searches your repository, edits files, runs commands, and verifies the change. Source-visible client.

Open-source, multi-tenant AI agent platform on AWS Bedrock — visual agent builder, RAG knowledge bases, evaluation suite, and omnichannel delivery (SDK widget, REST API, WhatsApp, Telegram). Self-hosted, MIT licensed.

Living Dream DSH - DeepSeek Harness desktop config framework with 8+ MCP servers, free model access, mobile remote, one-click install

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).

新一代（或许）跨平台、可扩展的 python 机器人平台 / multi-platform chatbot platform witten in Python

DRRM communication control center through AI-powered chatbots

Letta Code mod that detects non-English agent output (GLM Chinese-drift) and corrects it immediately — deterministic Unicode detection, turn_end + polling layers, anti-spam rails

DeepSeek Harness 批量并行子 Agent 调度插件：swarm_batch 工具（Kimi Code Swarm 模式移植）+ /swarm 命令 + 聊天内实时进度条

Multi-agent orchestration and LLM model routing for DeepSeek Harness (DSH): semantic AI agent profiles, exact model routes, declarative teams and strategies, and bounded subagent delegation - a TypeScript plugin that adds no second runtime.

GraphWorkflow is a compile-once engine that executes static multi-agent graphs with substantially lower orchestration overhead than LangGraph.

Autonomous reinforcement learning loop with adaptive hyperparameter tuning, multi-policy ensemble management, and framework-agnostic agent support. Self-converging via exponential backoff. Works with TensorFlow, PyTorch, JAX - bring any algorithm.

Multi-agent collaboration suite for DeepSeek Harness: specialist roster with on-demand dispatch, roundtable, model comparison and a multimodal vision bridge — models via the official provider flow.

Evidence-bound multi-agent council for Cursor and Claude Code. Scout (last30days), Advocate vs Skeptic, Fact Checker (deep-research). Ranked verdict with dissent.

TAC is not a language for writing code. It is a language for thinking like an agent. (The TacFlow Agentic Code) is a domain-specific language (DSL) designed for autonomous AI agents within the TacFlow swarm ecosystem. It is not intended for general-purpose programming.

A Rust-powered desktop SSH terminal and server-ops tool — multi-tab, SFTP, remote editor, session recording, Docker analysis & AI assistant

Production-shaped agentic AI customer support platform: LangGraph + Gemini agent, RAG over Chroma, MCP tool calling, JWT auth, RAGAS eval dashboard, Prometheus/Grafana observability.

Autonomous trend detection to pump.fun token launcher. Ten sources, independence-weighted scoring, hard-capped spend rails, a self-tuning agent inside a code-enforced allowlist, and a public dashboard that shows its own duds.

ArchGraph — an architecture-graph driven framework for Agentic Engineering. A long-term memory for coding agents, built on one ArchiMate 3.2 intent graph read/written through a single MCP interface.

AI数学建模竞赛助手 | 输入题目、自动生成论文框架+建模思路+Python代码+LaTeX模板 | MCM/ICM/CUMCM | DeepSeek驱动

Benchmarking GraphRAG and AriGraph on MuSiQue, HotpotQA and BabiLong QA1, run locally with Ollama and Llama 3.

基于孙宇晨长期公开表达形成的战略分析师，帮助普通人重新审视创业、财富、风险与人生选择。

Curated AI Agents, Automation & Engineering resources — each with a concrete first step on how to use it

Enclave — self-hosted, private legal AI for law firms & in-house teams. AI co-counsel, contract review, due diligence, drafting & playbooks, all inside your own VPC. Built on Onyx + Paperclip.

Memory Operations Platform for AI agents — ingestion, lifecycle, token-aware retrieval, and control UI

A personal AI assistant that runs entirely on your own machine: encrypted knowledge base, approval-gated tools, local models (MLX/Ollama) or your own cloud keys, built-in voice, and end-to-end-encrypted phone access. Source-available (ELv2).

some tool i decided to make which makes ai type for you in discord

微信聊天机器人。基于微信 iLink Bot API 的 AI 聊天机器人，支持 Termux 手机部署。内置 CoreCoder Agent 可执行文件读写、Shell 命令，流式分段回复模拟真人打字，网页控制台管理配置与日志。所有消息由 AI 驱动，可帮忙写代码、改文件、运行任务。

Community plugins for DeepSeek Harness (DSH) Web GUI: token cost meter, billing balance, update checker — plain JavaScript, no build required.

A local AI mock interview app that simulates real interview stages, supports DeepSeek/Qwen/ChatGPT/Claude, and generates detailed feedback reports.

Feishu bot bridge plugin for DeepSeek Harness

面向 Codex 的桌宠生成与校验 Skill，可创建、修复、验证、安装和打包 Desktop Pet Launcher 高清宠物包。

documentation for intentlink.io

2026 Agent 春晚 — 全球首届 AI Agent 春节联欢晚会：AI Agent 上台表演，配 CCTV-1 直播窗口与红包奖池实时大屏。中英双语。

A modular AI agent framework built on Vercel AI SDK. Zero-config agents with real tools, durable workflows, and HTTP interfaces.

15 AI agent skills for OpenClaw/ClawHub — sports-odds, kelly-criterion, portfolio-rebalancer, market-sentiment, streak-tracker, and more.

Terminal-first AI financial agent — runs fully offline via Ollama, 19+ cloud LLM providers, live market data, and a built-in quant research engine.

DeepSeek Harness plugin that auto-nudges agents to consult docs or web search when local trial-and-error becomes costly. DeepSeek Harness 插件：当本地试错成本升高时，自动注入检索提示，引导 Agent 查阅文档或联网搜索。

Trusted AI mathematics workbench: ProblemContract to Attempt to verified proof/counterexample bundles with SymPy and Lean evidence gates.

Lossless context folding for the Pi coding agent: only the oldest context changes; the fresh window stays untouched.

MCP server for Windows desktop automation. Incremental perception instead of a screenshot per step: measured 16.9x fewer tokens per action than the alternatives, with the benchmarks to check it.

An unofficial native macOS (SwiftUI) client for dsh, the DeepSeek agent harness — sessions, live trajectory, tool approvals, models, goals, subagents.

One MCP server for 185+ live web-data APIs: search, social, e-commerce, real estate, jobs, travel, news and finance. Clean JSON from sites that block scrapers.

A zero-dependency, offline-first visual UI for crafting production-ready JSON prompts for Gemini Image models. Replace messy text strings with precise, structured JSON. Features real-time schema validation and dynamic conflict warnings. Perfect for developers automating AI text-to-image pipelines.

In-game coding agent that autonomously plans, builds and self-inspects Minecraft structures (custom MCP bridge, deterministic generators, machine validators)

Open source AI voice calling agent for Twilio phone calls, built with FastAPI, Google ADK, and Gemini Live API

Yuj is a coding-agent harness that watches the LLM and keeps it on course. It maintains relevant context without extra LLM calls.

Open-source, serverless, multi-LLM group chat application

极境守护‑基于树莓派的多模态边缘安全守护系统

Multi-agent AI skill orchestration — 13 skills, 7 lifecycle phases, from idea to post-release. Works with Claude Code, Codex, Gemini CLI, Cursor, Windsurf, OpenCode, and any agent that reads Markdown or speaks MCP.

A wall of mail slots for coding agents running in parallel. Each git worktree gets a pigeonhole; agents post what they are working on and leave each other notes. No server, no daemon, no MCP. One POSIX shell script and the filesystem.

Engineering-grade investment decision system: deterministic engines + a multi-agent LLM debate engine

agent for qemu modeling build

Bedside: manners for AI agents that operate tools for smart, high-judgment non-experts. Ship and test, not only write.

AI-powered dashboard for global cyber threats, trends, and defense.

利用 Python 製作 Facebook Chatbot

Max gateway adapter for Hermes Agent

Disciplined task pipeline for AI agents: SPEC interrogation, self-contained tickets, proof-based execution, replan, retro. EN/RU docs.

Open-source voice agent platform with visible reasoning. Build a roster of AI experts with their own retrieval, guardrails, rules and RAG sources - configured from the UI, with per-step model routing across cloud providers or fully local via Ollama.

本地 AI 互动叙事引擎：多角色剧本杀流式叙事、结构化记忆、世界书注入、实况生图，整局存档可导出为加密 PNG「GG 卡」

An experimental configuration layer for intent-driven development.

Stateful agentic AI workflows on Amazon Aurora PostgreSQL with MCP, Amazon Bedrock AgentCore, Strands Agents, pgvector and row-level security

Open framework for building an agentic SEO app for Shopify — audits, JSON-LD automation, AI-search readiness. MIT.

[中文化]

Remote For Xirp — mobile web control surface for the Xirp agent daemon: machines, projects, sessions, live tmux pane. Unofficial.

考研 AI 学习助手 —— 基于 AI 的智能学习规划与复盘桌面客户端（Tauri 2 · Vue 3）

ChatBI Studio - governed open-source ChatBI and NL2SQL platform with verifiable results, semantic models, RAG, and bounded agents.

1,716 AI/ML interview questions with answer frameworks, 69 architecture diagrams, role-based study paths, and a voice-enabled mock interview simulator. Covers LLMs, RAG, agents, MCP/A2A, system design, MLOps, safety and cloud deployment.

Architectural showcase of Adam, a 3D-printed embodied AI humanoid integrating Google Gemini, Llama, and computer vision for ultra-low latency physical interaction.

AWS Bedrock RAG tutorial: Build a Q&A agent with Knowledge Bases and S3 Vectors (preview). Includes CDK deployment, web UI, and debugging tools.

20 free production-ready domain skills for Claude Code, Antigravity & AI agents — LangGraph, RAG, GEO/AEO, MCP, security

A 股量化交易工作台 — Java 21 + Spring Boot + Next.js 15 + FastAPI AI Agent + AI 投研問答（ToolCalling + MCP）的 Web 量化交易平台

📚 Interactive RAG knowledge base — 25 course videos + 20 advanced topics + production cheatsheet

Java + Python 企业知识库 RAG 工程项目：混合检索、可回归评估、有界并发与公网演示

Local-first AI fanfiction writing system with structured context engineering. Keeps your characters consistent across 100+ chapters. 本地优先的 AI 同人写作系统。

Form-filling agent built on browser-use: describe yourself once, point it at any web form, and it fills every field automatically. You review and submit manually.

AI agent skill for generating structured academic paper reading notes (Claude Code / Cursor / Codex)

Agent 的个性化 Agent：面向Windows 10/11，连接Codex及ACP Agent，提供桌面角色、认知、监督、陪看、情绪反馈和可扩展能力路由。

AI工程纪律框架——从6次真实事故中长出来的29条规则。不是提示词模板，是让AI少犯错、犯了错能被抓住、同样的错不犯第二次的纪律系统。

Weaving Stanzas with Wisdom 以智识 巧织诗

Open-source e-commerce AI image generation platform with multimodal RAG, AI agents, bestseller retrieval, product photography, quality evaluation, and batch workflows.

A runtime for reliable multi-agent systems with durable workflows, resumable execution, and human-in-the-loop control

Node.js Discord bot with voice conversation (wake-word, STT/TTS), OpenRouter-backed AI chat with tool calling, moderation, persistent memory, AI music generation, and a web dashboard.

Automatic isolated secondary review for Pi coding agent sessions

AI writing assistant plugin for DeepSeek Harness — draft and revise in chat, every edit staged on a rice-paper panel until you commit · 在 DSH 里使用青简：对话里起草改稿，右侧宣纸面板逐条审阅后落稿

本地优先、Agent 友好的中文创作工作台：把 Markdown 变成图文卡片和长文，支持 Codex CLI 与可选云端协作。

A verification-first Codex plugin for mathematical modeling competitions: problem analysis, reproducible modeling, rigorous figures, evidence-bound papers, and final audits.

Battle-tested patterns from running AI agents in production daily: the CLAUDE.md that finishes work, a 5-layer agent memory stack, overnight autoresearch loops, and LLM-judge alignment.

MacBook-sized decoder-only Transformer trained in an executable Sokoban RL environment

Deterministic investment decision system — discipline as code, every decision traceable and backtestable.

team-task for DeepSeek Harness (dsh): long-horizon multi-agent tasks — reviewed plan DAG, runtime-owned settlement, event-log truth, resident reconciler