Choose the foundations for an agent application: model integrations, orchestration and the developer tools around them. Your team's language, workflow complexity and operating environment are useful starting points.
Agency Swarm is a framework designed to automate AI agencies by creating a swarm of collaborative agents with customizable roles and functionalities, aiming to simplify the agent creation process and make automation more intuitive
A minimalist TypeScript middleware framework for building AI agents with composable guards, model routing, memory, testing, and observability.
Agent Reach installs and checks internet access tools for AI agents, routing search and platform access through supported CLIs and services.
A runtime for executing reusable agent skills with backend-agnostic capability contracts, policy and safety controls, and traceability.
A plugin layer for DeepSeek Harness that routes intent to tools, gates unused tool schemas, and adds persistent memory, event-triggered reflexes, and recovery behavior.
Run multiple coding agents in parallel inside isolated local Docker containers or cloud virtual machines, with per-box browser and VS Code access.
An open-source control plane for building and operating AI agents as callable services, with routing, coordination, asynchronous execution, memory, and observability.
The Agent OS is an experimental platform for creating self-evolving, autonomous AI agents capable of writing and executing their own code, designed to be a long-term environment for such agents and supports various programming languages
Agents Squads is a CLI for defining and running autonomous AI squads with persistent memory, shared goals, git-backed state, and verification across engineering, research, product, and operations.
AGiXT is an advanced AI Automation Platform designed to enhance AI instruction management and task execution across various providers, incorporating features like adaptive memory, smart instruct, and a versatile plugin system to push the boundaries of AI technology towards achieving Artificial General Intelligence (AGI)
Agno is a model-agnostic framework for building multi-agent systems with memory, knowledge, and reasoning.
A local workspace for coordinating AI coding agents across tickets, repositories, and pull requests, with project tracking, retained context, and human approval gates.
AIOS by AGI Research is an LLM Agent Operating System which enables an operating system 'with soul' -- an important step towards AGI
AIWG deploys specialized agents, skills, context, rules, and governed workflows into native project paths across multiple AI coding platforms.
A DSPy-native runtime that turns natural-language goals into inspectable agent graphs, executes independent steps in parallel, evaluates results, and can repair failed runs.
A multi-tool API and MCP server for AI agents, with x402 micropayments and tools for web, AI, crypto, and media workflows.
AutoGen is a multi-agent conversation framework facilitating the development of next-gen LLM applications, highlighted by various accomplishments and offering enhanced LLM inferences, customizable agents, and comprehensive documentation
AutoGen Studio 2.0 is Microsoft's advanced AI development tool, offering a user-friendly interface, powerful Python API, and comprehensive features for creating and controlling AI agents and workflows
Turn a shortlist of agent frameworks into a practical decision using task boundaries, recovery tests, integration needs and maintenance costs.
Read the guideSeparate task state, user preferences and source documents, then choose storage, retrieval and deletion rules for each type of information.
Read the guide