Custom agents

Save and use agent configs via sqlite

Try the live picker: https://trickaay.github.io/i-have-neurodivergence/ — Choose clearer, shorter AI replies. Optional skills for Claude Code, Codex and other assistants.

Entity-State-Transition-Constraint runtime for production-grade AI agents.

OSS Dependency Intelligence. Watch what can change underneath your software — EOL, license, distribution, support and security events with evidence..

Fully agentic offensive-security console. Plans, delegates, watches, revises, records, and reports. It has a production TUI, an experimental macOS GPUI shell, and very little interest in pretending that "AI" is a substitute for instrumentation.

Classify web traffic from server log entries and header capture into human, AI agent, AI crawler, AI assistant, programmatic, and other bot categories. No analytics service or middleware required.

An intelligent hybrid RAG platform powered by LangGraph + SGLang that seamlessly integrates knowledge retrieval with tool execution via MCP, featuring dynamic 3-way routing, quality-gate backtracking, and Graph-of-Thought reasoning.

Always-on embodied AI terminal for the home. A Raspberry Pi that sees you, knows who you are, and answers out loud through a lip-synced 3D avatar. Fully self-hosted, no cloud.

A skill that turns engineering and operational detail into writing a senior leader can act on in under a minute.

AGI OS is an Arch-based Linux distribution with AI agents built in — from installation to everyday use.

🤖 🌸 Personal Agent Protocol — Python and TypeScript SDK.

AI-agent ready (MCP): ask Claude or Cursor to audit your GTM container and Google Ads conversions, with evidence for every finding. Also runs in the browser (files never uploaded) and as a CLI. Read-only.

Local-first Windows AI assistant — voice wake word, long-term memory, OS automation. Runs on Ollama or any cloud LLM; your data stays on your PC.

Self-host the Hermes Agent as a Telegram bot (+ optional web dashboard) on a Blaxel sandbox. Cross-platform setup wizard (Windows/macOS/Linux), any model provider, always-on or scale-to-zero.

An AI coworker for team group chats, built on a pi-style terminal agent and TypeScript SDK.

Persian AI is a free and Open Source Chatbot ai and Model Manager

Learn AI architecture by building with a coding agent: hands-on labs for RAG, multi-agent systems and MCP servers, plus AI Center of Excellence playbooks.

Memgraph storage backends (KV, Vector, DocStatus) for LightRAG

AI assistant for a university chemistry library: grounded RAG chat + structured extraction

Text-blind context control plane for AI agents. Select high-utility context under token budgets.

Search index in Rust: documents in, ranked results out. Full-text, vector, and hybrid retrieval.

Gemini-powered IT helpdesk with FAQ routing, tickets, admin workflow, and local knowledge base.

AI-ready pentest/bug-bounty scenario dataset: generalized condition/step/impact/remediation playbooks grounded in real, publicly disclosed reports and writeups.

AI-powered resume analyzer — recruiter-grade ATS score, keyword gaps, 95%+ optimized rewrite, and a tailored cover letter. Built with Next.js + Google Gemini.

🏆 二等奖作品｜基于 HiAgent、多智能体协作与 RAG 的社区治理诉求闭环系统

本地化医学文献 RAG 问答系统。覆盖毕业论文文献调研、科研综述撰写、临床指南查阅、药学信息查询等场景 ，同时具备小数据量下的生产级可用性。支持多篇中英文医学论文 PDF（糖尿病、心血管、BMI 等）的智能检索与生成式问答。基于 Faiss 语义向量 + BM25 关键词 + Cross-Encoder 重排序三重检索 pipeline，模块化设计可自由切换嵌入模型、生成模型与重排序模型。全程不依赖任何第三方 API，基于 Ollama 本地部署 bge-m3 嵌入模型与 qwen2:7b 生成模型，确保医学数据隐私安全。实现了多会话隔离、流式 SSE 输出、异步 PDF 上传、前端历史持久化等生产级特性。

Open-source long-term, time-aware memory engine for AI agents (MCP + hybrid search + bi-temporal recall)

Multilingual Semantic Movie Search Engine using FastAPI, FAISS, and Sentence-Transformers (RAG Architecture)

A smart job radar that scores roles against a user's resume

One conversation for you and your AI agents: Claude Code and Codex discuss, build and check work together. Local-first, uses your CLI sign-ins.

Self-evolving agent framework on DeepSeek Harness (dsh): 40 Cordis plugins, 4 presets, memory · reflection · gates · metrics · an auditable evolution loop.

A unified simulator for fair comparison of agent memory systems

1000-class Chinese vehicle recognition with ResNet-50, deployed to K230D edge board (28 FPS offline) + ReAct agent

Hermes Agent (NousResearch) + Web UI as a Home Assistant add-on. Pre-baked, in-place upstream updates, HA Ingress, setup terminal.

把 Muse 接进微信的个人 AI 助手桥接：支持文字、图片、微信语音、视频和文件附件；可识图、转写语音、分析视频抽样画面、读写 Excel/Word/PDF/PPTX，并把生成图片和文件发回微信。内置资料找回、提醒待办、长任务进度、个人偏好和家人独立连接。Linux 自部署，需可用 Muse/muse-cli 与相应工具。PolyForm 非商业许可；内置 SDK MIT。

Local web workstation for the pi coding agent: resident sessions, live tool streams, PWA on your phone.

你来写，AI 管账：写长篇小说的桌面软件，逐章记账、核前后矛盾、设定本，数据全在本机 | Write your novel, let the AI keep the books: per-chapter ledger, continuity checks, story bible, 100% local

Deploy DeepSeek-V4.1 (dsv4.1) on Ascend 910B / 910B3 (NPU): a hardware-specialized LLM inference engine with W4A8, DSpark speculative decoding, 128K+ context and 8,700+ tok/s prefill. Also supports NVIDIA A100/A800. 昇腾910B 部署 DeepSeek-V4.1-Flash。

Agent-ready PHP component framework for admin dashboards. Zero dependencies, Material Design 3, AJAX-first. 16 components, 1 CSS + 1 JS file.

生产环境调试神器 - AI 驱动的运维助手，支持日志分析、进程排查、Docker/K8s 诊断，单文件部署，即插即用

Agent Skills 统一管理工具 —— 跨平台统一管理 AI 编程工具的 Skills 技能包

🎬 Model Context Protocol (MCP) server & Agent Skills for CapCut Desktop & JianYing Pro. Automated video editing, multi-track timeline construction, Quad Ease keyframing, and viral typography for Claude, Cursor, and Antigravity.

Independent Android app for Hermes Agent by Nous Research: mobile remote gateway, streaming chat, sessions, files and private Tailscale access. Flutter client with APK downloads.

Auto-Apply That Ends on an Interview. GetMattered MCP server: job matches, applications sent on the employer's own hiring system with the confirmation they return, per-role resumes, cover letters and interview prep. Remote MCP at mcp.getmattered.com

Claude Code plugin for professional furniture design — AI skill for structural validation, cut optimization, BOM generation, and FreeCAD 3D visualization.

AxiomCore — a personal project & routine management agent for Claude. Enforced structure, plan→approve→execute discipline, wiki memory, optional voice/brand/todos/dev-planning modules.

从内网出海：内网在中国大陆的公司集群里，AI Agent 及其沙箱怎样稳定访问海外服务。写给不懂网络的程序员的小书

Same stock, dual wrappers, live gap. Real-time tokenized US stock price comparison, reference gap analysis, and best-execution routing across bStocks and Ondo on BNB Smart Chain.

Reverse-engineer any website into a clean Next.js codebase — powered by SuperDuck browser automation.

The mobile client for opencode. Offline-capable, with incremental sync and a local copy of your sessions.

Multi-brand social content AI agent — short-form video pipelines, ORM with hard guardrails, multi-platform publishing. Open source (MIT).

An AI agent for open mathematical problems — honest by design.

TRACE — AI-native HPC workbench for supervised research workflows, schedulers, resilient transfers and reproducible execution

Zero-data-loss file snapshotting, selective rollback, and token-optimized audit trail for AI coding agents (Cursor, Claude Code, Cline, Windsurf, Antigravity).

A practical Tasfia AI persona prompt for authorized cybersecurity learning, defensive research, CTFs, and secure development.

你的眼 YourLiveEye · Agent-friendly Bilibili live-chat analysis, history backfill and Cloudflare deployment

Secure MCP gateway that lets ChatGPT, Claude, Grok & other AI agents remotely code, run commands, manage Git, control browsers and desktop apps, operate multiple workspaces, and supervise your device/server from anywhere—with OAuth, approvals, sandboxing, DLP, audit logs & rollback.

An AI sparring partner for your mind. Project Eidolon stress-tests your ideas by refracting them through six intellectual domains—from physics to philosophy—including a 'Divergent' lens for unconventional thinking. It's a tool for thought, powered by a symbolic LLM-to-LLM protocol.

Run Your Own Private AI Agent on a Spare PC — a sanitized, hardware-agnostic guide to self-hosting the Nous Research Hermes Agent privately with Docker + Tailscale (CLI/TUI, persistent dashboard, nightly backups, remote desktop access, and an AI-agent setup prompt).

AI agent for analog and mixed-signal circuits: LTspice schematics and IC blocks on the IHP SG13G2 open PDK, sized at every corner, simulated on ngspice, Xyce, LTspice or Spectre, laid out, checked with DRC and LVS, and handed to ChipLeap as macros. Desktop app, CLI and MCP server.

Your PC won't start? Your phone takes over. An open-source AI rescue kit for broken computers, on a rooted Android phone and a USB cable.

A Wireshark-style packet viewer you and your AI agent use together, over MCP

养一条会自己长大的鱼：没有人格设置，人格只能通过事件流长出来。DSH 插件 · 三层记忆 · QQ / B站私信 / 抖音 三个真实渠道。

行记：可插拔城市的大模型旅行规划 Agent（Web 端 + MCP 服务）。一条命令接入新城市，已接北京、上海、广州、深圳、成都、重庆、武汉、杭州、南京、苏州十城；一句话说需求、规划前确认；LangGraph + OR-Tools 排程与规则校验。

Autonomous AI personal assistant on Telegram integrated with Google Calendar and Gmail for daily scheduling, task tracking, and planning.

IdeaTaste - A benchmark for research taste: let LLM agents propose research ideas, then check how many of them the research area actually went on to do.

Convergent deliberation loop for coding agents: multi-pass, coverage-gated, append-only ledger review and planning. Claude Code, Codex, OpenClaw.

Multi-agent Discord Gateway bridge for Grok Bot + Hermes — bot-peer allowlists, not blanket bot drops.

Deep Search Pro - 一个轻量的多智能体协作系统（Agent 开发入门实战项目）

Let Claude Code and Codex review each other's code from your current session. One-line install for macOS / Linux / WSL2 sets up the official CLIs if missing; uses your existing Claude / ChatGPT subscription.

Multi-account, multi-model AI collaboration workbench: roles backed by different AI accounts review each other, use leftover quota, and work across multiple servers.

多人多系统的 AI Coding 敏捷看板

An open-source pixel-art office where AI agents work as a team: editable roles, any model, routing by a lead, and PC tools with permissions.

Agent skill for exploring what-if scenarios through personas, multi-round swarm simulations and structured scenario reports inspired by MiroFish.

把你的 Hermes AI 员工放进一间看得见的办公室：老板办公室 / 经理办公室 / 开放工位区 / 聊天室 @ 派活

One manager agent, on-demand workers, fewer tokens: Claude Code plans, Codex/Claude/ccr workers build in isolated git worktrees, the manager verifies and merges.

Non-blocking agent workflows for pi: background jobs, durable timers, threads and git/jj sandboxes.

An OpenCode & Codex CLI agentic flow for OpenSpec, the most loved spec framework — plan with critique, apply in parallel waves, validated mechanically

PI Desktop 的 DeepSeek-Bot 独立适配版：长期 Bot 团队、群聊协作、记忆、Skill 与语音。An independent adaptation of DeepSeek-Bot for PI Desktop.

Orchestrate any mix of coding agents and models, from Claude Code and Codex to open-weight models on your own machine.

Task-coordination store for agent fleets — a dependency graph of beads in SQLite, one unblocked bead per request via an atomic single-transaction claim, and a git-tracked checkpoint for recovery. Rust.

🌍 A survival sandbox world for AI agents — they trade, cooperate, specialize, or die. Rust + Python + Next.js.

Offline-first fiction writing environment. Import a novel, let a local model read it into a story bible, then ask "What if...?": branch the book, rewrite the chapters the change reaches, and check every later scene for contradictions and skipped events. LangGraph multi-agent writing pipeline; Vue 3 + Dexie; Ollama.

Analytics engine for creators on X: explains why posts work, recommends what, how and when to post, and tests drafts on a simulated Argentine audience.

Continuity Plane｜面向长期 AI 辅助开发的 provider-neutral 控制面，解决压缩遗忘、重复工作与多 Session 竞态；reliable recovery, bounded context, evidence gates, checkpoints, and safe collaboration.

Lets Claude Code agents in different repos work together: each knows the others, proposes work to them and gets their answers back, with your approval for every step

omp-pstack brings Poteto's P Stack to Oh My Pi (OMP): multi-model AI coding agents, engineering playbooks, verification, and durable automation.

Open-source coordination backend for AI agent teams. Shared memory, messaging, tasks via MCP. 20 tables, 75 tools. Convex-native.

A generalized, AI-powered product deployment assistant. Uses a multi-agent pipeline (Strands + Amazon Bedrock) to guide users through deploying any product catalog on AWS: gathering requirements via interview, generating architecture designs, producing Infrastructure-as-Code (CloudFormation), and creating documentation.

A living pixel-art AI agent office / 会动的俯视像素 AI 办公室：自动分工、多 Agent 协作，支持 Ollama、Codex、Claude Code、Hermes 和自定义 API。

Multi-model AI agent stack: one model builds, another model lineage reviews. Runs Claude Code, Codex, Gemini, Grok, Kimi & more on your own CLI subscriptions. 60+ skills, task graphs, agent messaging, shared memory.

A native Rust workspace for parallel AI coding agents, with persistent terminal sessions and Git worktrees.

Distributed heterogeneous agent task dispatcher — claim under lease, return with evidence. Quota-aware routing across Claude, Codex, OpenCode Go and DeepSeek, in isolated git worktrees with hard budgets.

Research into a new method of context management

AutoApply AI is an **Agentic AI system** that automates the job application workflow. Instead of manually searching and applying for jobs, this AI agent:

Did it push straight to main? Touch a file you told it never to? RuleReceipt checks a finished AI-agent session against your CLAUDE.md / AGENTS.md rules and marks each followed, broken, or can't-tell — with the exact line. Runs locally, nothing uploaded.

Turn an existing website into a discoverable, reliable, and measurable agent interface. Delivered through native WebMCP.

Native Codex plugin and Hermes Agent skill kit for game studios working across Unity, C++,C#, Lua, MMORPG services, databases, production, release, and live operations.

A command-line interface toolkit that enables AI agents and automation scripts to drive Autodesk Revit through an HTTP API.

Assurance methodology for autonomous AI agents in regulated financial services