Custom agents

A modern AI chatbot frontend built with AlpineJS

iran seda crawler golang

融入团队的 AI 员工 · Self-hosted AI teammate platform connecting WeCom and Feishu/Lark to Claude Code and Codex runtimes you control.

Use your openclaw anywhere, web and mapp

Telegram‑бот на Python (async, python-telegram-bot v20+). Команды + автоответы по правилам, логика протестирована, токен в .env.

FoxyMoe is a discord bot written in nodejs with discord.js. This is a chat bot that communicates via the oobabooga webui API (LLM model running locally).

A local-first Windows AI desktop assistant. One-click launch, three-layer memory, pluggable tools. Runs on your machine — your data never leaves it. 本地优先的 Windows AI 桌面助手：一键启动，三层记忆，插件化工具，数据全在本机。

A Javascript plugin to help developers build simple conversational chat flow.

A modular, multiplatform chatbot with features for both fun and utility to enhance the chat experience.

Customizable Desktop Pet Companion

🤖 AI-Powered Course Recommendation Chatbot — Personalized learning path suggestions using Flask & IBM Watson NLP

闪客快打7重置版 (CFN) 的 RAG 智能对话后端：基于游戏资源做检索增强，支持与 NPC 流式对话、任务发布、好感度、立绘与知识库管理。

A self-hosted Discord bot that grows its own personality — background AI agents remember everyone, form opinions and pick their own name. Rust · any OpenRouter/Mistral model.

张雪峰风格的高考志愿AI助手

Self-hosted AI agent platform — durable memory, hybrid RAG, sandboxed tools & plugins, multi-channel chat (QQ / Telegram / Discord / Slack / Feishu / WeChat), and a web admin console with human-in-the-loop approvals. 自托管智能体平台：持久记忆、真实工具、多渠道接入、运维后台。

Starter kit for educators to test, adopt and iterate on genAI tools in the classroom

A self-hosted AI storytelling app, heavily inspired by RisuAI

Beauty-retail customer-service AI copilot — RAG hybrid retrieval, intent routing, compliance guardrails, grounded answers

记得住事、也会主动找你的陪伴型 Agent · Windows 桌面版 · 本地优先

General purpose virtual assistant with GUI and LLMs.

An LLM integrated Discord chatbot project, built & designed to provide automated assistance.

Self-hosted Instagram DM automation bot in Python and a free ManyChat alternative. Auto-replies to comments and sends follow-gated DM links via the Graph API on GitHub Actions cron.

Turn your herdr agents into bots that talk to each other — give Claude Code, Codex or Grok a name and a chat, and put a few in a room. macOS app + herdr plugin.

Your agent has agents. A self-hosted, local-first AI agent platform in one Go binary: agents hand work to specialists and build the tools, agents and apps they lack, with sandboxed tools and credentials the model never sees.

A personal agent built around a continuously maintained digital twin. Local context, specialist teams and lasting work.

Local-first team workbench for AI agents: project mailboxes, shared context, task handoffs and human review. v0.8.2 · macOS App and local Web; Intel, Windows and Linux previews.

Autonomous Goal Execution & Multi-Turn Task Tracking Engine with Sticky Header for DeepSeek Harness

Autonomous agentic platform for software development and other activities. Specialist agents, constitution-bound AI Employees, evidence-based execution, multi-transport RPC/HTTP/WebSocket, 5-tier memory, full self-improvement cycle. CLI + TUI + Flutter GUI + macOS MenuBar.

🎓 校园社团智能招新平台 | AI-Powered Campus Club Recruitment Platform — React 18 + Supabase + 豆包大模型 · 7 意图 AI 助手 · 中英双语

Free, open-source SSH/SFTP client for Windows with split terminals, Xshell import, ZMODEM transfers, and a customizable AI assistant.

Windows 微信 AI 助手：多会话人设、长期记忆、本地维护台与高级设置。Clone and reproduce with your AI.

Hannah, a local AI assistant with a voice, a 3D body and hands. Backend, avatar, desktop overlay and launchers.

A Flipper Zero agent on your iPhone. FlipperHero pairs over Bluetooth and lets an AI explore files, fire Sub-GHz and IR, emulate NFC, RFID and iButton, craft payloads and install apps, with every risky move gated behind your approval.

Free Agentic coding tool on a free ChatGPT account. A VS Code extension + browser extension that lets ChatGPT read your GitHub repo, apply its edits, check them and iterate — no API keys no payment.

Chief: replies from Grok Bot, ChatGPT (Codex), Claude, Gemini and Cursor on your Meta Ray-Ban Display glasses or any browser. Encrypted, with one-tap Yes, No, Later replies.

Enspira is a RAG-powered LLM interface, complete with reranking, summarization, classification, and TTS capabilities. While not documented too well at the moment, feel free to ask questions.

AI-powered chatbot for MkDocs — chat with your docs and get instant answers

An AI PDF Chatbot built with Retrieval-Augmented Generation (RAG) that lets you upload PDFs and ask questions using local embeddings (sentence-transformers), FAISS vector store, and the Claude API.

Multi-provider web search with automatic fallback, circuit breaker, and zero-config defaults. 8 built-in providers (DuckDuckGo, Brave, Bocha, Metaso, Tavily, Gemini, SerpAPI, SearXNG). Zero runtime dependencies.

AANU Jewellers, Thorrur: gold, silver and combination jewellery store with live rate ticker, order-time rate lock, 3D viewer, AI assistant (Groq), 10,000+ design catalogue and installable PWA.

Open-source, self-hostable harness for AI support agents on WhatsApp and the web: RAG, LangGraph agent, webhook tools, embeddable widget.

Leak Hunt: hands-on demo of bringing AI to your data. A .NET assistant, an egress traffic monitor, Foundry Local and a browser lab. Synthetic data only.

Grounded question answering over the EU AI Act, with retrieval, faithfulness and hallucination measured on 45 hand-written questions.

Turn any folder into an MCP knowledge server: local-first indexing, hybrid search, answers with span-level citations.

RAG platform for educational use at DSV, Stockholm University. Rust/Axum backend, React frontend, Qdrant vector search.

Australian location + scenario to reviewable preparedness drafts. Hybrid RAG, evidence trails and human review. Local Ollama; password-protected cloud demo. Not emergency decisions.

题库 PDF 进 · 随机抽题出 —— 非结构化题库 PDF 的结构化抽取 + 智能刷题系统（规则优先 + LLM 兜底 / 三池加权抽题 / Leitner 间隔重复 / 带引用讲解）

Map, graph and audit AWS, Azure and GCP from one CLI. Builds a topology graph, runs Prowler/Checkov/Trivy, tags findings against 28 compliance frameworks, and exports HTML reports, RDF ontologies, RAG chunks and Terraform recreation files.

BOB YouTube: Docker-first YouTube ingestion runtime for developers, products, bulk workflows, and AI agents. Powered by yt-dlp, wrapped as a clean CLI, OpenAPI service, /data corpus contract, and release-proof workflow.

MSc dissertation (GISMA, 2026): a controlled benchmark of indirect prompt injection, PoisonedRAG and Crescendo against four open-weight LLMs in a RAG pipeline, with three low-cost defenses, exact statistics and an EU AI Act Article 15 mapping.

Self-hosted semantic code search. Index repos and knowledge sources into a local vector DB, then search by natural language from CLI or any MCP-compatible agent.

Open, reproducible benchmark of US legal-answer quality. Verified questions, a standard-library scorer, and results anyone can rerun.

多人联机文字跑团 demo：Agent 担任 DM，RAG 约束剧本 + 剧情状态机 + 服务端权威骰子；FastAPI + WebSocket 大厅/开房间联机；三路模型通道（云 API / llama.cpp / 自研 C++ 内置引擎，CPU 与 CUDA 双后端）。

SuperNovel: An advanced AI-powered novel generation framework. Craft coherent, long-form narratives with multi-agent collaboration, dynamic plot structuring, and persistent world-building memory.

Install AI capabilities, not libraries. Versioned, reusable AI components with a dependency-free CLI.

Failure-oriented benchmark for PDF-to-Markdown, OCR, and VLM document parsers with executable assertions and public release leaderboards.

No-code AI chatbot platform with Zero-Speculation RAG. Train chatbots on your documents and embed them into any website in minutes.

Private in-browser editor for screenshots, photos and PDFs: font-matched text replacement, OCR, redaction, translation and code export. Designed and built by NIKHIL CHARY SRIRAMOJU.

边缘智能体管理平台 — 让工业网关拥有 AI 大脑 | ONNX 推理 · LLM · RAG · 多智能体编排 · 断网自治 · 全协议接入

This is a free, open-source course on Large Language Models (LLMs) — designed for everyone from complete beginners to developers who want to build production-grade AI apps.

A feature-rich desktop GUI for Ollama with Vision, RAG, and JSON support.

Benchmark contamination and clustered inference in rare-disease gene prioritisation — a 1,047-case benchmark with per-item leakage flags, model-free floors, and fully re-derivable results.

Pure TypeScript document parser for PDF → Markdown, JSON and RAG chunks.

基于 Qwen 大模型与 RAG 技术的中医药 AI 健康助手，提供体质辨识、AI 问诊、用药参考等功能

AI/ML engineer interview questions for 2026 — LLMs, RAG, fine-tuning, embeddings, agents, MLOps, prompt engineering, and system design for AI. Junior to staff level.

AdCraft AI - Hyper-Engine v2.0: A production-grade AI-powered advertising engine featuring a Neural Concept Engine, real RAG retrieval, CNN logo analysis, and a browser-based PWA interface.

Full-stack MTG Commander AI platform — rules-complete simulation (Forge + Monte Carlo), LLM deck builder with sim-driven feedback, RAG-powered coaching, ML training pipeline, LAN battlefield, and Unity client.

Self-hosted, permission-aware RAG for internal knowledge bases: hybrid search, rerank, cited answers, full traces. One process, one embedded database.

Chat with any website. Scrape a page, ground Gemini on its content, and have a persistent AI conversation. Built with Next.js, Prisma, Firecrawl, and PostgreSQL.

Unified MCP server for scraping Salesforce developer documentation and searching a local SQLite vector-like database.

Fragmentos de código del libro "Programar con Inteligencia Artificial" de Joaquín Ruiz (@jokioki). Ejemplos prácticos sobre código conversacional, agentes IA, MCP y flujos ▎ AI-first para developers. 📖 Libro en Amazon | 🎥 YouTube @jokioki

A craft knowledge base for prompting AI video toward looking real, not AI-made - 1000+ specific, checkable rules across cinematography, lighting, editing, and product design.

On-device AI camera lens: detects objects, explains what they are used for, and chats with Groq/Gemini. PWA with WebGPU, TensorFlow.js, voice and multilingual replies.

Document-to-knowledge engine: normalizes PDF, DOCX, XLSX, TXT and images into one versioned, source-linked model that AI assistants, apps and websites can use.

Ask and get answers with their sources, straight from the Micro Club's documents.

The core API and orchestration layer for FaultMaven, connecting the browser extension and dashboard to your infrastructure and LLMs.

Self-hosted personal search. Index and search across your files, email, chats, and scanned documents from one interface. Highlighted snippets, hybrid ranking, and optional semantic search.

Integrate Boogu-Image pipelines into ComfyUI with custom nodes for text-to-image, image editing, and fast inference.

Governed AI orchestration for Node.js — build, run and govern production AI agents with built-in access control, audit trail and human-in-the-loop.

AuraVector DB: High-precision, low-latency hybrid vector engine for RAG systems. Combines BGE dense embeddings, Rank-BM25, Reciprocal Rank Fusion (RRF), Cross-Encoder reranking, and SVD entropy analysis to eliminate LLM hallucinations under a p95 < 300 ms constraint. Built for ADROSONIC Build Challenge by Team SynaptiX (BIT Mesra).

吉康环境 · AI 智能市场推广创作工作流 —— 多智能体公众号推文自动创作系统

AI twin of NIKHIL CHARY SRIRAMOJU: talking avatar, RAG with a vector DB, LangGraph agents, voice, emotion detection, JWT auth. FastAPI + Docker backend, GitHub Pages frontend.

Three measured experiments on RAG hallucination: quote-checking, TypeSafe's Jev, and IBM's STAIR. 850+ graded questions, raw responses included.

Self-hosted Agent quality engineering platform — evaluation as a release gate. 8 eval types with AI judges, datasets, pre-eval scripts, analytics & A/B testing. Gate loop (API keys, CI CLI, webhooks) lands with M1. FastAPI + Vue 3 + Celery.

An unofficial, non-commercial, read-only MCP server that lets LLM hosts use Mapeak (formerly Israel Hiking Map) hiking data.

Logos is a tool which allows all your ai agents to share the same context and pick up right after one another

A clicker for LLMs: operate web pages and desktop screens via shell commands.

Official CapMonster Cloud MCP server — AI captcha solver for reCAPTCHA v2/v3, Cloudflare Turnstile & DataDome. Let Claude, Cursor, and other AI agents solve captchas directly via MCP.

A freelancer's whole agency, run by 19 AI agents: they find clients, build their website, pitch, call, quote, plan and chase payments for you. You just approve. Showcase repo (code is private).

Judge-level court filing rules, court holidays, and privacy enforcement data for US courts, as an MCP server. Hosted at mcp.courtrules.app

DeepSeek Harness 的上下文压缩插件：自动与手动压缩、工具结果裁剪、有界溢出恢复。

全双工语音助手：边听边说、打断、任务委派与工具执行。Full-duplex voice agent with barge-in, streaming audio, and async thinking. DuplexOmni (arXiv:2606.09186).

Agent-first TypeScript library for safe SaaS API connectors: typed actions, pagination, retries, rate limits, and LLM tools. MIT.

🌱 Context Garden is a context management suite for coding agents on multi-repo platforms. Shares system knowledge across repositories and agent swarms, and cuts what each agent reads: exact code ranges, clustered logs, remembered repo facts. Offline Claude Code skills, no LLM calls.

Ironheights: The Immune System for AI Agents. Protects any AI bot that installs it: scans agent skills (OpenClaw, Claude, Grok Bot, Codex, Cursor, MCP) A-F, threat-model PR checks, MCP server, drift guard, posture report. Local, no telemetry.

面向 Android 与 Windows 的多模型 AI Agent 工作台，支持流式对话、工具调用、项目开发、MCP 与逐次审批。

AI 长期推理决策 Benchmark｜中转站 URL+API Key 全自动跑模型、实时日志、排行榜、赛后复盘与模型画像

Local-first AI workspace for long-form fiction: outlines, world-building, Memory Bank, drafts, AI review, rewrites, and confirmed-chapter gates.