kepano/obsidian-skills
Obsidian Skills provides a collection of agent skills enabling AI agents to interact with Obsidian vaults, facilitating content creation and editing of Markdown, Bases, and JSON Canvas files, as we...
Awesome AI Agents › AI Agent Development
AI Agent Development
Panniantong/Agent-Reach
Agent-Reach allows AI agents to access web content from major social media and video platforms through a CLI, without API fees, enhancing their perceptual capabilities.
JCodesMore/ai-website-cloner-template
This project provides a reusable Next.js template for reverse-engineering and cloning any website into a modern codebase using various AI coding agents.
TencentCloud/TencentDB-Agent-Memory
TencentDB Agent Memory provides a local, 4-tier progressive pipeline for AI agent long-term memory, enhancing performance and reducing token usage without external API dependencies.
huggingface/smolagents
Smolagents is a minimalistic Python library by Hugging Face for creating intelligent agents that think and act by generating and executing Python code, supporting multiple LLMs, modalities, and tool integrations.
alibaba/page-agent
Page Agent is a JavaScript in-page GUI agent enabling natural language control of web interfaces without browser extensions or headless browsers, supporting various LLMs.
titanwings/distilly
A meta-skill that distills messages, documents, interviews and public sources about a person into a source-grounded Person Profile, packaged as an Agent Skill for coding agents and chat bots.
KKKKhazix/khazix-skills
Khazix Skills is an open-source collection of AI agent skills, including tools for disk analysis, AI news retrieval, project documentation tidying, and report generation, compliant with the Agent S...
coze-dev/coze-studio
Coze Studio is an all-in-one AI agent development platform that simplifies the creation, debugging, and deployment of AI agents through visual tools and supports no-code and low-code development approaches.
Alibaba-NLP/DeepResearch
Tongyi DeepResearch is an advanced open-source large language model by Alibaba designed for deep, long-horizon information-seeking tasks with state-of-the-art agentic search performance.
browser-use/browser-harness
Browser Harness is a self-healing CDP-based browser automation tool that connects LLMs directly to a real browser, allowing agents to perform tasks and generate reusable skills dynamically.
microsoft/agent-lightning
Agent Lightning is a flexible server-client framework that enables training and optimization of AI agents using reinforcement learning and other algorithms with minimal code changes.
transitive-bullshit/agentic
Agentic is a versatile AI agent standard library that integrates with any large language model and TypeScript AI SDK, providing a comprehensive set of AI functions and tools for building intelligent applications.
opensandbox-group/OpenSandbox
OpenSandbox is a general-purpose, secure, and extensible sandbox runtime for AI applications, offering multi-language SDKs and robust isolation for agent execution.
pinpoint-apm/pinpoint
Pinpoint is an APM tool for large-scale distributed systems that provides real-time monitoring, transaction tracing, and detailed performance analysis with minimal impact on application performance.
TencentCloud/CubeSandbox
Cube Sandbox is a secure, high-performance sandbox service built on RustVMM and KVM, designed for AI agents with instant startup, hardware-level isolation, and E2B compatibility.
bytedance/trae-agent
Trae Agent is an LLM-based agent with a modular, research-friendly architecture designed to execute complex software engineering tasks via a natural language CLI interface using multiple LLM providers.
OpenPipe/ART
OpenPipe/ART is an open-source reinforcement learning framework that trains multi-step agents for real-world tasks using GRPO and innovative zero-shot reward scoring with LLMs.
strands-agents/harness-sdk
Strands Agents is a versatile SDK for building and operating AI agents with model-driven approaches, supporting various LLMs, multi-agent systems, and production deployments.
superradcompany/microsandbox
Microsandbox provides local-first, microVM-backed programmable sandboxes for AI agents, offering lightweight, hardware-isolated, and instantly-booting environments.
google/adk-go
An open-source Go toolkit for building, evaluating, and deploying sophisticated AI agents with flexibility, modularity, and cloud-native support.
MiroMindAI/MiroThinker
MiroThinker is an open-source research agent model designed for advanced tool-augmented reasoning and complex information-seeking tasks, achieving state-of-the-art performance across multiple benchmarks.
op7418/guizang-social-card-skill
This is a Claude Code/Codex skill for generating social media carousels and cover pairs with editorial and Swiss design aesthetics, supporting various aspect ratios and visual styles.
plastic-labs/honcho
Honcho is a memory infrastructure for building stateful AI agents that understand and adapt to changing entities over time, offering reasoning-first memory and multi-peer perspectives.
steipete/agent-scripts
This repository hosts shared instructions, skills, and portable helper scripts for AI agents, streamlining development and deployment across multiple projects.
NVIDIA-NeMo/Guardrails
NeMo Guardrails is an open-source toolkit by NVIDIA for adding programmable guardrails to large language model-based conversational systems to ensure safer, controlled, and trustworthy interactions.
agent-infra/sandbox
agent-infra/sandbox is an all-in-one sandbox environment combining browser, shell, file system, MCP services, and VSCode Server in a single Docker container to provide a unified, secure, and versatile platform for AI agents and developers.
apache/maka
Local-first Apache Incubator agent workspace that runs coding-agent tool calls under a sandbox and records every turn as a durable execution log.
areal-project/AReaL
AReaL is an asynchronous reinforcement learning infrastructure designed for scalable training of large-scale reasoning and agentic models, bridging foundation model training with agent-based applic...
coze-dev/coze-loop
Coze Loop is an open-source AI agent optimization platform providing full-lifecycle management from prompt development and debugging to evaluation and monitoring, enabling efficient AI agent development and operation.
TaskingAI/TaskingAI
TaskingAI is an open-source BaaS platform for developing, managing, and deploying AI-native applications using large language models with unified APIs, customizable tools, and multi-tenant support.
pguso/ai-agents-from-scratch
AI Agents From Scratch is an educational repository that teaches users to build AI agents locally using local LLMs and node-llama-cpp, focusing on understanding agent architectures, function calling, memory, and the ReAct pattern through progressive, hands-on examples.
Agenta-AI/agenta
Open-source workspace for building agents through chat using AGENTS.md, skills and MCP servers, with swappable harnesses, background runs, per-tool permissions, tracing and cost tracking.
FareedKhan-dev/all-agentic-architectures
All Agentic Architectures is a comprehensive repository offering practical implementations of over 17 advanced AI agentic architectures, designed to educate and empower developers in building intelligent, adaptive, and collaborative AI systems.
CaviraOSS/OpenMemory
OpenMemory is an open-source, self-hosted long-term memory system that enhances AI by providing persistent, multi-sector, and explainable memory capabilities with high performance and user isolation.
JoshuaC215/agent-service-toolkit
AI Agent Service Toolkit is a comprehensive framework for building, deploying, and interacting with AI agents using LangGraph, FastAPI, and Streamlit, featuring multi-agent support, streaming, content moderation, and a user-friendly chat interface.
zai-org/GLM-4.5
GLM-4.5 is an open-source large language model designed for intelligent agents, featuring hybrid reasoning capabilities and high performance across industry benchmarks.
google-antigravity/antigravity-sdk-python
Python SDK from Google for building agents on Antigravity and Gemini, abstracting the agentic loop behind stateful conversation objects with streaming, multimodal input and read-only defaults.
memodb-io/Acontext
Acontext is a context data platform for building cloud-native AI agents that store, observe, and learn from contexts and tasks to improve agent performance and scalability.
Intelligent-Internet/ii-agent
II-Agent is an open-source framework that builds and deploys intelligent agents leveraging leading large language models to automate complex workflows and enhance productivity across multiple domains.
Meirtz/Awesome-Context-Engineering
Awesome Context Engineering is a comprehensive survey and resource repository focused on the evolution from prompt engineering to dynamic, context-aware AI systems, providing theoretical foundations, methodologies, and practical guides for large language models and AI agents.
MiroMindAI/MiroFlow
MiroFlow is an open-source, high-performance research agent framework designed for building and scaling intelligent AI agents to perform complex tasks like future event prediction with state-of-the-art results.
MemMachine/MemMachine
MemMachine is an open-source universal memory layer that enables AI agents to store, recall, and manage multiple types of memory to create personalized, context-aware AI assistants for various applications.
MiniMax-AI/Mini-Agent
Minimal reference agent implementation in Python showing how to build agents on the MiniMax M2.5 model, with a full tool loop, session memory, context summarization, MCP and Claude Skills.
FullAgent/fulling
Fulling is an AI-powered full-stack development platform that automates the provisioning, coding, and deployment of full-stack applications using a sandboxed environment with Next.js, Claude Code, PostgreSQL, and Kubernetes.
jackmpcollins/magentic
Magentic is a Python library that enables seamless integration of Large Language Models into Python code by using decorators to create functions with structured LLM outputs, supporting advanced features like streaming, function calling, and observability.
UnicomAI/wanwu
Wanwu is an enterprise-grade, multi-tenant AI agent development platform by China Unicom that supports intelligent agents, workflows, model management, and integrates large language models with business process automation for digital transformation and business innovation.
langwatch/better-agents
CLI tool and project convention from LangWatch that scaffolds agent repositories with scenario tests, evaluation notebooks, versioned prompts, MCP config and an AGENTS.md for coding assistants.
oceanbase/powermem
PowerContext (PowerMem 2.0) gives coding agents durable project-scoped memory through a local SQLite-backed server with a Python client, Core SDK, CLI and Codex and DeepSeek Harness plugins.
photon-hq/imessage-kit
TypeScript iMessage SDK for macOS that reads, sends and watches conversations from the local Messages database, intended as a communication channel for AI agents and chat-first automation.
OpenHands/software-agent-sdk
Python and REST SDK from OpenHands for building agents that work with code, running either on the local machine or inside ephemeral Docker and Kubernetes workspaces through an agent server.
DataDog/dd-agent
Datadog/dd-agent is the legacy version 5 of the Datadog monitoring agent that collects and reports system and application metrics to the Datadog platform, supporting custom checks and configurations.
mpaepper/llm_agents
A small library to build intelligent agents controlled by large language models, enabling task-solving through tool integration in a Thought-Action-Observation loop.
vercel-labs/knowledge-agent-template
Vercel Labs template for building knowledge-base agents that search sources with grep, find and cat inside sandboxes instead of embeddings, and deploy as a web chat, GitHub bot or Discord bot.
LLaVA-VL/LLaVA-Plus-Codebase
LLaVA-Plus is a project developing large language and vision assistants that learn to use tools for enhanced multimodal AI capabilities.
ruvnet/metaharness
MetaHarness is a TypeScript generator that scaffolds a custom agent harness for a given repository, emitting a branded npx CLI, a local MCP server, scoped memory, generated skills, governance policy and signed release provenance for hosts such as Claude Code and Codex.
Chen-zexi/open-ptc-agent
Open PTC Agent is an open implementation of programmatic tool calling: the agent writes Python that calls MCP tools inside a Daytona sandbox, so only final results return to the model context.
mnemon-dev/mnemon
Mnemon is a single Go binary that gives CLI coding agents persistent cross-session memory: the host LLM decides what to remember and how to link it, while the binary handles a four-graph store with recall, decay and deduplication.
strands-agents/sdk-typescript
Strands Agents TypeScript SDK is a model-driven agent framework for Node.js and browsers with Zod-typed tools, MCP support, streaming and graph/swarm multi-agent orchestration; now archived into harness-sdk.
truffle-ai/dexto
Dexto is a configuration-driven agent harness in TypeScript that supplies orchestration, sessions, memory, tools and observability for building AI agents, and ships with a ready-made coding agent.
Ingenimax/agent-sdk-go
Ingenimax/agent-sdk-go is a powerful Go framework for building production-ready AI agents with multi-model support, advanced memory management, modular tools, and enterprise features.
genlayerlabs/yeagerai-agent
Yeager.ai Agent is a flexible and interactive tool for creating, prototyping, and deploying AI-powered agents within the Langchain ecosystem, featuring session persistent memory and an interactive CLI.
deepset-ai/haystack-cookbook
The Haystack Cookbook is a collection of example notebooks demonstrating the use of the Haystack AI framework for various search and retrieval tasks, aimed at helping users learn and implement Haystack effectively.
strands-agents/agent-builder
Interactive command-line toolkit built on the Strands SDK for creating, testing and extending custom agents and tools, with hot-reloading, more than a dozen built-in tools, nested agents, swarms and Bedrock knowledge bases.
blob42/Instrukt
Instrukt is a terminal-based integrated AI environment that enables users to create, instruct, and manage modular AI agents securely within sandboxed containers, featuring document indexing, tool attachment, and a rich terminal interface for efficient AI workflows.
krohling/bondai
BondAI is a comprehensive platform for building and deploying highly capable single and multi-agent AI systems with extensive integrations and easy-to-use interfaces.
sematext/sematext-agent-docker
Sematext Docker Agent is a deprecated tool for collecting host and container metrics, logs, and events, with users directed to updated Sematext monitoring solutions for Docker, Kubernetes, and infrastructure.
phact/streaming-assistants
A deprecated library that enabled streaming for OpenAI Assistants API using Astra Assistants, now replaced by the astra-assistants package.