Custom agents

Evidence-bound assurance for agentic AI — governance, provenance, authorization, audit, and fail-closed lifecycle controls.

Evidence-governed multi-agent decision system with blind expert rounds, red teaming, RAG, decision memory and typed verdicts.

Resident peer agents for pi — long-running partners inside your session that tick on their own rhythm, watch the work, and push findings back into the conversation.

GAIA — Generative AI Interface for Agents: one plans and decides the WHAT, specialist subagents execute the HOW — with project memory, context, techniques and verification built in — and an execution gate that deterministically classifies each command, escalating to you only what really matters.

Agentic planning for Vision-Language-Action robotics — orchestrating SmolVLA/LeRobot policies with LangGraph for multi-step robot manipulation.

Build durable agent harnesses so smaller models can do repeatable, verifiable work — a versioned folder with tools, guardrails, validators, and measured evidence from every run.

A modern, terminal-based dashboard for managing tmux workspaces. Built with Python Textual, it features a command palette, live previews, robust session persistence, and seamless zero-friction resumption for AI coding agents.

"Too Many Tools" CLI coding agent.

19,000+ Agentic AI APIs, AI model APIs, autonomous agents, and MCP servers for building AI agents, copilots, automation, and LLM apps.

GDG Build With AI Event - Building Agentic AI using google ADK

THE UNIVERSAL AGENT SKILLS LIBRARY

Web UI for Hooman: chat with your agent, manage MCP servers and skills, and tune configuration—built with Next.js, React, and TypeScript.

ai-agent java

Structured context for AI agents

Implementation of Chatbot using Intent Classification.

Send Facebook Messenger and Instagram messages from Laravel — every message type, with pre-flight validation and clear, structured errors.

Context compression for LLM pipelines that proves nothing protected was lost before it emits, and refuses when it cannot. Refusals count against the published numbers. Paper, benchmarks and receipts ship with the code. Open core by LaconIQ. Apache-2.0.

A self-evolving multi-agent learning system with learner profiling, RAG, quality evaluation, reflection, debate, and adaptive learning paths.

早期RAG后端：文档解析、分块、Embedding与向量检索

[ACL 2026] The official GitHub repository for DFAMS: Dynamic-flow guided Federated Alignment based Multi-prototype Search

AI-powered placement intelligence for students. Analyze your resume, match job descriptions, identify skill gaps, and get a personalized roadmap.

🚂 Query your files locally

Self-hosted cross-grounding Q&A for code and documents, with verifiable evidence.

Understand long videos by decoupling perception and reasoning through hierarchical graph memory and an agentic retrieval mechanism.

Zero-hallucination citation engine for multi-document research. NotebookLM-style grounding in your terminal.

Transform uncertainty into absolute confidence.

🛡️ Open-source AI security scanner & LLM red-teaming platform. Test LLM APIs, chatbots, agents, MCP servers & RAG for prompt injection, jailbreaks, data leaks & unsafe tool use — with OWASP LLM Top 10 mapping and plain-English, audit-ready reports.

AWS-native AI Security Posture Management for AI agents (Observe • Govern • Defend). Discovery, OWASP/NIST/MITRE posture rules, runtime detection & Bedrock Guardrails enforcement, and AIDR integrations (Security Hub, GuardDuty) — open source, in-account, extensible.

A video editor an AI agent drives, running on your own machine.

Python 3.14+ backend capability framework for applications and AI agents. Protocol-neutral business operations with explicit HTTP and MCP adapters, machine-readable contracts, policies and invocation lifecycle.

lightweight, high-performance Personal AI Agent assistant workspace built using Node.js

Real-time interview and meeting assistant: captures system audio, transcribes with Groq Whisper, detects questions, and streams AI answers to an overlay hidden from screen capture. Runs on one free API key.

[EMNLP2026] "A Dual-Detection Multi-Agent System for Paper-Code Discrepancy Detection""

A zbus-style message bus that keeps external integrations out of the OpenHuman kernel's dependency graph.

A practical open-source roadmap to build production-grade AI agents, copilots, RAG systems, MCP tools, and agentic applications.

Multi-agent system that autonomously optimizes vLLM GPU kernels for a specific deployment — model, hardware, dtype, parallelism — over multi-hour unattended campaigns

A native VS Code client for DeepSeek Harness (DSH) with sessions, streaming replies, tools, model controls, and workspace-aware workflows.

🦉SmartOwl is an app that helps you prepare for exams by creating personalized study resources from your own learning materials.

PlotRail companion skill for existing manuscript intake, diagnosis, refinement, style learning, continuity audit, and compact handoff.

CloserAI - a local-first, model-agnostic, permission-transparent desktop AI workbench built on DeepSeek Harness.

DeepSeek Harness 的 AdaptiveContextPlane (ACP) 插件——带治理的长期记忆系统

Learn how GRC (Governance, Risk & Compliance) skills are driving modern careers. Includes salary insights, certifications, and top global learning resources.

AI 足迹 — 跨平台智能收藏管理工具。AI自动分类整理、共享协作、Agent API接入。支持PC/手机H5/浏览器扩展。

Cross-agent marketing strategy pack with 29 Agent Skills, 24 specialist agents, neural marketing graph, evidence challenge gates, and deterministic utilities.

TriSoul —— 三魂共识 Agent：三个灵魂盲写+匿名互评，三官（对齐/博识/实证）补偿生成，画布式上下文，记忆中枢。DeepSeek Harness 插件套件，一键装/独立端口/一键卸载

Turn your AI agent into a white-label marketing agency — daily research, ideas, leads, videos, blogs, publishing, and analytics for every client.

Autonomous Red vs. Blue AI Fraud Simulation & Defense System for UPI & Digital Payments. Features LLM-driven zero-day attack generation (F3 Taxonomy), PyTorch Bidirectional LSTM + LightGBM + GraphSAGE multi-model defense, and real-time SIEM dashboard.

Rust multi-agent coordination platform: HTTP/SSE gateway, agent runtime with token-budgeted context, per-tool sandboxing, multi-provider LLM support.

Agentic memory-augmented retrieval and evidence grounding for medical question-answering tasks

ARI is a production-grade, open-source AI voice assistant and agentic orchestration platform. Powered by FastAPI, Go Fiber, React & WebGL, ARI delivers real-time voice command processing, intent detection, multi-step LLM planning, and localized telemetry optimization.

An autonomous AI agent that lives inside your real Telegram account over MTProto — not a bot. Reads and searches your chats, acts under a permission policy you control, answers people while you're away, and takes instructions from any chat. Sandboxed code execution, prompt-injection defence, installable plugins.

Software workshop focused on agentic AI, exploring AI agents, tools, reasoning, memory, autonomous workflows, and practical development through introductory experiments.

Agent skills for Cursor, Claude Code, and Codex — reliable agent loops with scope, verify, and resume

Agentic Vision-Language Assistant for Earth observation. SatQuery AI fuses Optical and SAR satellite imagery using a deterministic LLM router for visual grounding, bi-temporal change detection, and VQA through natural language.

Catch AI-hallucinated package names before you install them

An agent-first calendar for an AI companion's own social life, projects, rituals, and temporal orientation.

AI-powered resume + DOCX template autofiller using local LLMs, Flask, and ChromaDB (RAG-based)

Local-first RAG with an incremental ontology over your Markdown, exposed to any AI agent via MCP.

From-scratch audio-AI research system + 99-lesson bilingual course — DSP · ASR · Music · LLM/RAG, every core hand-written, no API wrappers.

Reusable, tool-agnostic AI engineering skills, workflows, templates, and evaluation playbooks for AI coding agents.

AI support agent that handles tier-1 questions, detects frustration, and escalates to humans before customers churn. LangGraph + FastAPI + pgvector.

Your friendly neighborhood QA developer.

gbrain 可视化界面

用 Codex 从 PDF 简历生成可对话、可自部署的专属面试 Agent 网站

Automated machine learning pipeline using MLflow, DVC, Docker, GitHub Actions, and AWS deployment for reproducible AI workflows.

Claude Code skill for building and auditing Android/Jetpack Compose/KMP apps — NowInAndroid architecture + a 10-dimension code audit & refactoring workflow

AI Research Radar — Evidence-Grounded Trend Ledger for AI Systems

Claude Fable 5 orchestrator for React Native/Expo — Grok 4.5 Heavy implements; Argent UI verify; STATUS + ledger multi-agent chains

A software factory of Claude Code sessions driven by GitHub issues, labels and pull requests

Autonomous software building powered by Codex, Archon workflows, and Agent Orchestrator.

This is a FastAPI LangGraph+ReAct Agent with chat-based news+weather inquiries

Claude Code plugin for persistent AI memory. Store conclusions, not notes. OODA-driven reasoning artifacts with priority matrix, invalidation triggers, and seed activation.

TaskSwarm (蜂群) — DeepSeek Harness 上的多智能体任务编排插件：waves/lanes 并行执行、git worktree 隔离、任务包与跨模型评审、崩溃可恢复 | Multi-agent task orchestration for DeepSeek Harness: waves/lanes parallel execution, git worktree isolation, task packets, cross-model review, crash recovery

Саморазворачивающийся мультиагентный фреймворк для Claude Code: Facilitator + dev-команда subagent-ролей, персистентная память, гейты качества

基于 LangGraph、FastAPI 与 Vue 3 的 AI 智能导购助手，支持实时商品检索、多智能体分析、价格与口碑对比、连续追问及报告导出。

An intelligent research paper assistant that uses Retrieval-Augmented Generation (RAG) to help users search, understand, and interact with research papers through contextual question answering

Shared knowledge for AI teams. The opinionated starter kit for a shared knowledge canon — built on Obsidian, Git, and Markdown. Every session makes the team smarter.

Minimal agent that runs in pi on exe.dev VMs

Diversio Agent Skills marketplace using the open standard; includes Claude marketplace metadata

Legora — independent third-party profile of a public API surface, by API Evangelist. Legora is a Stockholm- and London-based legal AI company (formerly Leya) building a collaborative AI workspace for lawyers. The platform is centered on Legora aOS, an agentic operating system that connects information, communication, and the execution of legal work

TabMail iOS client

Hermes Agent 的专属桌面伴侣 — 桌面宠物 | AI 记忆系统 | 智能工具箱 | 语音交互 | 一键启动

Agent Harness 学习指南与面试题库 — 零基础到系统理解，73 张图解 + 37 道题 + Pi/dsh 架构深潜（中文）

Codex plugin for durable agentic software-factory orchestration with SQLite-backed ledgers.

Self-hosted agent for infrastructure

YOLO-Agent is an evidence-driven autonomous optimization harness for object detection | YOLO-Agent 是一个面向目标检测的证据驱动自主优化 Harness

75 production-grade agent skills for Hermes Agent + Paperclip — research, write, organize, earn, and run an AI workforce. Every skill passes a QA gate with hard safety rails. Built by Gen AI Hub.

Security-first agent harness built for long horizon research and work capabilities with increased focus on math, computer science, and domain science research

Simplest Trained AI Chat Bot with Google Drive and Apps Script

A 24/7 automated RAG agent for answering finance policy questions. Features an n8n no-code pipeline and a Python shadow layer with FastAPI and Streamlit.

A Go LINE Bot demo showcasing message edit and unsend webhooks through a group-ordering story with live Flex Message summaries.

Feishu and Lark text and image channel plugin for DeepSeek Harness

A reusable, domain-neutral SwiftUI chatbot for iOS, featuring Apple Foundation Models, configurable knowledge sources, pluggable providers, persistent conversations, comprehensive error handling, and accessible, customizable UI.

基于 DeepSeek-V3 的混合检索增强生成 (Hybrid RAG) 科研助手。支持本地向量库 ChromaDB、Rerank 重排序及联网搜索。

CodeMentorAI is an AI-powered codebase mentor that helps users analyze, understand, and interact with GitHub repositories using retrieval-augmented generation, repository summarization, architecture insights, bug detection, and conversational code Q&A built on Streamlit and Groq.

🧠 A 7-layer cognitive memory system for AI agents — with sleep cycles. Not another vector store.

A local RAG-based document question-answering system using FastAPI, FAISS, Sentence Transformers, and Gemma 3:1B.

面向视频内容解析与理解的RAG问答平台

The SQLite of AI Memory — free, open-source, zero-dependency Python library for persistent LLM memory. Works with Claude, GPT, Gemini, Ollama, Cursor, and any MCP client.

A complete Retrieval-Augmented Generation (RAG) pipeline designed for Vietnamese documents. This project implements a full system architecture from document ingestion and hybrid retrieval to grounded generation and automated evaluation, utilizing local LLMs and async operations.