AI Agents

Agents and agent frameworks we track, from coding assistants to autonomous research tools. Each entry covers what it does, what it integrates with, and where it has been evaluated.

178 entries

• Deep AgentsLangChainOpen-source, production-oriented agent harness with planning, subagents, filesystem tools, persistent memory, and human approval controls.
• PydanticAIPydanticTyped, extensible Python SDK for building production-grade AI agents.
• Claude Agent SDKAnthropicSDK for building production agents with Claude Code capabilities for codebases, tools, files, and terminal workflows.
• AutoGenMicrosoftOpen-source framework for building agentic AI applications with multi-agent communication, tool use, and code execution.
• Cursor AgentAnysphereCursor Agent is a coding agent for inspecting repositories, editing files, running terminal commands, and iterating on software engineering tasks.
• Vercel AI SDKVercelTypeScript toolkit for building AI-powered apps and agents with tool calling, streaming, model providers, workflows, and frontend-friendly abstractions.
• pestPestPHPPest is an elegant PHP testing framework with Composer installation, a documented CLI test runner, and official agent-oriented docs for AI coding-agent test verification.
• Google Agent Development Kit (ADK)GoogleOpen-source framework for building, evaluating, and deploying AI agents with strong Gemini and Google Cloud integration.
• OpenHands CloudOpenHandsHosted cloud version of OpenHands for delegating software engineering tasks to AI coding agents.
• GitHub Copilot coding agentGitHubGitHub Copilot’s autonomous coding agent works on GitHub issues, makes changes in a development environment, and opens pull requests.
• Gemini CLIGoogleOpen-source command-line AI agent from Google that brings Gemini models into the terminal for coding, automation, and local project-context workflows.
• OpenAI Codex CLIOpenAIOpenAI’s lightweight coding agent for the terminal, designed to work locally with a developer’s codebase.
• Augment AgentAugment CodeSoftware-development agent in Augment that plans and implements end-to-end coding tasks, edits code, runs tools, reviews diffs, and supports Agent Auto.
• DroidFactory AIFactory’s agent-native software development platform and CLI coding agent for planning, implementing, reviewing, and automating engineering work.
• Loki ModeAutonomiSpec-driven autonomous coding agent that turns PRDs, GitHub issues, OpenAPI files, or briefs into verified software with quality gates.
• ProtoAgentThomas GauvinMinimal TypeScript AI coding agent CLI with tool-use loops, approvals, sessions, MCP, skills, sub-agents, and a build-your-own tutorial.
• Agent Swarmdesplega-aiOpen-source, MCP-powered framework for orchestrating teams of autonomous AI coding agents with lead-worker delegation, Docker isolation, GitHub integrations, and PR-oriented delivery.
• CoderClawBuilderforce.aiSelf-hosted open-source AI coding assistant and multi-agent orchestration platform for developer workflows and codebase automation.
• OpenAgentask-solOpen-source terminal-native AI coding agent with multi-provider support, local sessions, command execution, file edits, web search, and MCP support.
• OpenJetL-ForsterLocal-first terminal coding agent that sets up local LLM backends, reads repos, edits code, and runs approval-gated commands on your machine.
• SWE-agentSWE-agentOpen-source coding agent that takes GitHub issues and attempts to automatically fix them using a chosen language model.
• WrongStackWrongStackTerminal AI coding agent with REPL, TUI, and web UI surfaces for reading and editing code, running commands, and using subagents, skills, tools, and multiple providers.
• Refact.ai AgentSmall Magellanic Cloud AI Ltd.Open-source autonomous AI coding agent for IDEs with repository context, tool integrations, self-hosting, and on-prem enterprise deployment options.
• KaitmoretonLocal open-source AI coding assistant with persistent memory, tool use, background agents, CLI and Web UI access, skills, and git workflow support.
• HivemootHivemootOpen-source AI engineering-team framework that runs agents locally and coordinates GitHub issues, PRs, reviews, votes, and shipping workflows.
• OptioJon WigginsOpen-source orchestration system for AI coding agents that drives tickets to pull requests, monitors CI/reviews, and resumes agents through feedback loops.
• agent-boberBOBER3rAutonomous multi-agent coding pipeline with planner, builder, and evaluator roles for building and verifying software features.
• yo, gomdmagnuson-creatorAgent system for opencode with specialized agents, skills, scaffolds, and project templates for multi-project software development workflows.
• Coder AgentsCoderSelf-hosted AI coding agent built into Coder for delegating software work on controlled infrastructure with governance, auditability, and model choice.
• TalkCodyTalkCodyFree, open-source AI coding agent with local storage, multi-provider models, parallel agent workflows, skills, and native desktop performance.
• AgenticjbcomPolyglot AI agent orchestration toolkit for coding-agent fleets, triage, PR review, crew workflows, and sandboxed multi-agent execution.
• PR-AgentQodoOpen-source pull request agent for automated code review, PR description generation, suggestions, and Git provider workflows.
• LocalCodelsheaselAutonomous terminal AI coding agent that runs locally with Ollama or LM Studio, edits code, runs commands, checks diagnostics, and manages git.
• Zed AgentZed IndustriesNative AI coding agent in the Zed editor for reading, editing, navigating, and running code in developer projects.
• CodeWhaleHmbownOpen-source terminal-native coding agent with local execution, approval gates, rollback, sub-agents, sessions, tools, and multi-provider model routing.
• AutoCodeRoverAutoCodeRoverSG / SonarAutonomous software engineering agent for resolving GitHub issues, bug fixes, and feature additions using structure-aware code search and patch generation.
• gdmguidegdmOpen-source terminal-native AI coding agent with whole-codebase reasoning, autonomous edits, local model support, autonomy controls, and audit logs.
• ShogoShogo LabsOpen-source agent platform with an agent-native IDE and CLI for writing, refactoring, reviewing, and shipping code in repositories.
• Hexihexi-codesMinimal contract-driven runtime and CLI for bounded coding-agent steps against local git repositories, with replaceable adapters and structured event logs.
• CodebuffCodebuffTerminal-based AI coding agent that indexes a codebase and coordinates agents to edit, refactor, and generate code from natural-language instructions.
• ContinueContinue Dev, Inc.Open-source AI coding agent for CLI, VS Code, JetBrains, and CI workflows.
• Roo CodeRoo Code Inc.Open-source AI coding agent for VS Code with specialized modes, model-agnostic providers, code editing, terminal use, and project-aware workflows.
• JulepJulep AIOpen-source platform for building persistent AI agents and long-running multi-step agent workflows with memory, branching, tools, and APIs.
• SwivalSwivalOpen-source CLI coding agent that works with frontier, local, and open models, with context management, review loops, memory, and sandboxing options.
• BMad MethodBMad CodeOpen-source AI-native agile development framework with specialized agents, workflows, skills, and lifecycle guidance for software teams.
• Open SWELangChainOpen-source asynchronous coding agent framework for internal engineering agents with sandboxes, Slack/Linear/GitHub triggers, subagents, and PR creation.
• MotiaMotiaDevCode-first backend framework for building APIs, background jobs, workflows, automations, and AI-agent-oriented systems across TypeScript, JavaScript, and Python.
• Agentic Ops FrameworkAgentic DevOpsYAML-native AI agent framework for DevOps, SRE, PR review, incident response, and ops automation.
• Trae AgentByteDanceOpen-source CLI agent for general-purpose software engineering tasks, with multi-LLM support, file editing, bash tools, and trajectory recording.
• Qwen-AgentQwenLMOpen-source agent framework for Qwen-based LLM applications with planning, tool use, memory, RAG, code interpreter, GUI demos, and custom agent development.
• AI SDKVercelProvider-agnostic TypeScript toolkit for building AI-powered applications and agents across React, Next.js, Vue, Svelte, Node.js, and more.
• JunieJetBrainsAI coding agent by JetBrains for JetBrains IDEs, Android Studio, terminal, GitHub Actions, and GitLab CI/CD.
• KaibanJSKaiban AIJavaScript-native framework for building and managing multi-agent systems with a Kanban-inspired approach, task flow, and state management.
• OpenSeekOpenSeek contributorsOpen-source terminal coding agent with multi-provider routing, MCP, LSP feedback, and Plan/Agent/YOLO autonomy modes.
• ShepShep AIOpen-source local-first orchestrator for running multiple coding agents in isolated git worktrees from feature prompt to PR.
• ZilFluentDataOpen-source CLI and Python SDK for validating, packaging, evaluating, signing, and deploying production AI agents as portable artifacts.
• CosineCosineAI coding agent for real software engineering work across CLI, cloud, and desktop.
• gptmegptmeOpen-source terminal AI agent for coding and automation with shell, Python, web, vision, patching, plugins, local-first use, and persistent autonomous agents.
• aiagentflowaiagentflow contributorsLocal-first CLI multi-agent software engineering orchestrator using Architect, Coder, Reviewer, Tester, Fixer, and Judge agents.
• hermes-studioEKKOLearnAISelf-hosted AI chat dashboard for Hermes Agent with multi-platform chat, session management, scheduled jobs, and usage analytics.
• BlitzyBlitzyAutonomous software development platform that maps large codebases, orchestrates specialized agents, generates validated code, and opens pull requests.
• IBM BobIBMEnterprise AI software development partner for SDLC workflows, repository understanding, security checks, modernization, coding, testing, and deployment assistance.
• MonkMonkAutonomous DevOps agent that connects coding agents to deployment, infrastructure provisioning, operations, diagnostics, and production runtime workflows.
• OpenCastleMonkiLabsOpen-source multi-agent orchestration framework for coordinating coding assistants into specialist software development teams.
• CodiveCodiveOpen-source terminal AI coding agent for understanding codebases, editing files, running commands, searching code, and working with existing editors across multiple LLM providers.
• Tyrex FrameworkTyrex FrameworkAgent-agnostic workflow framework for coding agents with structure, project memory, guardrails, slash commands, and sub-agent orchestration.
• harnextQualityUnitTerminal-native AI coding agent harness for reading and editing files, running shell commands, driving MCP servers, and automating coding tasks.
• CodeboltCodeboltAI agent engine for building software with CLI, desktop editor, cloud workspaces, SDKs, custom agents, and multi-agent orchestration.
• PlandexPlandex AIOpen-source terminal AI coding agent for large, multi-step software tasks across real-world codebases.
• DifyLangGeniusOpen-source, production-ready platform for building agentic workflows, AI agents, RAG pipelines, and LLM apps with visual tooling and APIs.
• LangroidLangroidPython framework for multi-agent LLM programming with Agent and Task abstractions, tool/function calling, RAG, and orchestration.
• Kilo CodeKilo Code, Inc.Open-source agentic coding platform for VS Code, JetBrains, CLI, Slack, and cloud agents.
• SwarmClawSwarmClawOpen-source, self-hosted AI agent runtime and multi-agent framework for autonomous agent swarms with memory, delegation, schedules, and coding-agent integrations.
• KaganMakerXKeyboard-first Kanban TUI that orchestrates coding agents across task planning, execution, review, and merge workflows.
• Warp AgentsWarpAI coding agents integrated into Warp’s agentic development environment, supporting local and cloud agent workflows from the terminal.
• QoderQoderAgentic coding platform for real software development with task delegation, agentic chat, browser agent support, and IDE integrations.
• DevinCognitionAutonomous AI software engineer for coding tasks, bug fixes, PRs, codebase Q&A, testing, and SDLC workflows across cloud, CLI, and desktop surfaces.
• JulesGoogleAutonomous coding agent that connects to GitHub repos, plans tasks, works in a Cloud VM, produces diffs, and can create pull requests after review.
• Letta CodeLettaMemory-first coding agent and SDK for persistent agents that learn over time, work locally or remotely, and can be embedded into agentic coding applications.
• GitHub Copilot cloud agentGitHubAutonomous GitHub Copilot agent that works in a GitHub Actions-powered environment to research repositories, change code, and open pull requests.
• AutomakerAutoMaker-OrgLocal open-source agentic IDE that uses Kanban tasks and autonomous developer agents to plan, write, test, fix, and review software changes.
• mini-SWE-agentSWE-agentMinimal open-source software engineering agent for solving GitHub issues or local command-line coding tasks with a small, hackable Python control loop.
• Google AntigravityGoogleAgentic development platform and CLI for managing autonomous coding agents, parallel local agents, background subagents, and software workflows.
• AG2AG2Open-source framework for building AI agents and enabling multi-agent collaboration with LLMs, tools, human-in-the-loop workflows, and conversation patterns.
• AgentCrewSaigon TechnologyMulti-agent AI assistant framework for coding, research, and automation with desktop, terminal, API, and headless job-runner modes.
• gooseAgentic AI FoundationOpen-source local AI agent with desktop, CLI, and API surfaces for code, workflows, automation, MCP extensions, and multi-provider LLM use.
• AgentOSFrame.devOpen-source TypeScript AI agent framework with cognitive memory, multi-agent orchestration, guardrails, tool forging, and multiple LLM providers.
• BoringOSHebbsOpen-source agent framework and shell for running coding agents as CLI subprocesses with tasks, workflows, memory, approvals, budgets, and audit trails.
• OrlojOrlojHQOpen-source orchestration runtime for multi-agent AI systems, using YAML to define agents, tools, policies, workflows, scheduling, and governance.
• CodexOpenAIOpenAI coding agent for building, refactoring, testing, reviewing code, and running local terminal workflows through the Codex CLI and ChatGPT surfaces.
• CodeepVlado IvankovićAutonomous terminal AI coding agent that reads projects, edits files, runs commands, and verifies changes with user-selected models.
• Windsurf CascadeWindsurfCascade is Windsurf’s agentic AI coding assistant for AI-powered development workflows.
• LlamaIndexLlamaIndexFramework for building LLM-powered agents and workflows over private or external data, with tools, memory, RAG, and multi-agent patterns.
• Replit AgentReplitAI software-building agent inside Replit that can plan, write code, set up infrastructure, test, iterate, and deploy apps from natural-language prompts.
• Microsoft Agent FrameworkMicrosoftOpen Python and .NET framework for building, orchestrating, and deploying production AI agents and multi-agent workflows.
• DroidsFactoryAI coding agents for planning, writing, testing, and shipping code from natural-language tasks.
• Cursor Background AgentsAnysphereAsynchronous coding agents in Cursor that run in remote environments to edit code, run commands, report status, accept follow-ups, and support takeover.
• KiroAmazon Web ServicesAgentic development environment and terminal coding agent for spec-driven software work, hooks, steering, chat, code edits, commands, and pull requests.
• Dapr AgentsDaprFramework for building durable, resilient, observable AI agent systems with workflow orchestration, statefulness, security, and telemetry.
• Strands AgentsStrands Agents / AWSOpen-source Python and TypeScript SDK for building production AI agents with tools, hooks, multi-agent patterns, observability, and broad model-provider support.
• AmpSourcegraphFrontier coding agent for terminal and editor workflows, with CLI, subagents, thread sharing, plugins, and multi-model coding modes.
• Docker AgentDockerDocker Agent is Docker’s open-source framework and runtime for building teams of specialized AI agents with declarative configs, packaged and distributed through Docker workflows.
• AgentScopeAgentScope AIProduction-ready agent framework for building, running, observing, and serving AI agents with tools, permissions, workspaces, and agent services.
• OpenCodeAnomalyOpen-source AI coding agent for terminal, IDE, and desktop workflows with LSP support, multi-session agents, and broad model-provider support.
• AiderAider-AITerminal-based AI pair-programming agent that edits existing codebases, maps repositories, runs lint/tests, and integrates with Git.
• AgnoAgnoOpen-source Python framework and platform for building, running, and managing production AI agents and multi-agent systems.
• Claude CodeAnthropicAgentic coding system from Anthropic that reads codebases, edits files, runs commands and tests, and works across terminal, IDE, desktop, web, Slack, and CI/CD integrations.
• ClineCline Bot Inc.Open-source coding agent runtime for IDEs, terminal, SDK, Kanban, and CI workflows with file editing, command execution, tool use, and human approvals.
• MetaGPTFoundationAgentsMulti-agent framework that models a software company with PM, architect, engineer, and other roles to turn requirements into specs, APIs, docs, and code.
• VoltAgentVoltAgentOpen-source TypeScript AI agent framework and engineering platform with agents, workflows, memory, tools, MCP, observability, evals, and guardrails.
• agentic-project-managementAgentic Project ManagementAPM is a framework for managing complex projects with spec-driven multi-agent workflows.
• smolagentsHugging FaceLightweight open-source Python library from Hugging Face for building agentic systems, especially code-executing agents.
• OpenHandsAll Hands AIOpen-source platform and SDK for autonomous software development agents that can plan, edit code, run tests, and open PRs across real codebases.
• LangGraphLangChainLow-level orchestration framework and runtime for building, managing, and deploying long-running, stateful AI agents and multi-agent workflows.
• Agent Development KitGoogleGoogle’s open-source, code-first framework for building, debugging, evaluating, and deploying production AI agents across multiple languages.
• harborHarbor FrameworkFramework for specifying sandboxed agent tasks for evaluation and optimization.
• AgentsMeshAgentsMeshAI Agent Workforce Platform for orchestrating AI coding agents at scale from a single platform.
• boxliteBoxLiteEmbeddable, stateful micro-VM sandboxing for AI agents with snapshots and hardware isolation.
• bitterbot-desktopBitterbot-AIOpen-source desktop AI agent focused on local-first operation, persistent memory, emotional intelligence, and a peer-to-peer skills economy.
• OpenSandboxOpenSandboxUniversal sandbox infrastructure for AI applications, described as a secure, fast, and extensible sandbox runtime for AI agents.
• wanwuUnicomAIChina Unicom's Yuanjing Wanwu Agent Platform is an enterprise-grade, multi-tenant AI agent development platform for building agents, workflows, RAG applications, and managing models.
• kiteKite ContributorsKite is a modern Kubernetes dashboard that brings observability, multi-cluster operations, user governance, and AI agents into one workspace.
• rl-baselines3-zooDLR-RMTraining framework for Stable Baselines3 reinforcement-learning agents with scripts for training, evaluation, hyperparameter tuning, plotting, videos, and pretrained agents.
• clawpanelqingchencloudClawPanel is an OpenClaw and Hermes Agent multi-engine AI management panel with a built-in AI assistant, multimodal support, and a Tauri v2 cross-platform desktop app.
• mercury-agentcosmicstack-labsSoul-driven AI agent with permission-hardened tools, token budgets, and multi-channel access for CLI or Telegram operation.
• openlitOpenLITOpen-source AI engineering platform for monitoring, debugging, and improving LLM applications with observability, tracing, evaluations, prompt management, guardrails, GPU monitoring, and related tools.
• telegram-searchgroupultraExport and fuzzy search Telegram chat history, with documented semantic search, vector search, and RAG-powered AI Q&A features.
• nexunexu-ioOpen-source Electron desktop client for connecting OpenClaw agents to IM channels such as Feishu, Slack, Discord, and WeChat.
• MiroFlowMiroMindAIMiroFlow is an agentic AI platform for building intelligent agents with tool integration, benchmark evaluations, and a web UI.
• go-microGo MicroPluggable Go framework for distributed systems with service discovery, RPC, pub/sub, and AI agent integration via MCP.
• openai-agents-jsOpenAIOfficial OpenAI TypeScript SDK for building agentic AI apps, multi-agent workflows, and voice agents.
• agency-swarmVRSENOpen-source agent orchestration framework for building reliable multi-agent systems with the OpenAI Agents SDK and Responses API.
• WebwrightMicrosoftTerminal-native web-agent harness where models use Playwright code and bash commands for long-horizon web tasks.
• koogJetBrainsOpen-source JetBrains JVM-native framework for building production-ready AI agents in Kotlin or Java.
• LazyLLMLazyAGILazyLLM is an all-in-one tool for building and optimizing multi-agent LLM applications.
• claude-devtoolsmatt1398Free, open-source DevTools-style debugger for Claude Code sessions.
• open-slide1weihoopen-slide is a React-first slide framework built for agents.
• PromptWizardMicrosoftOpen-source framework from Microsoft for automated prompt and example optimization using feedback-driven critique and synthesis.
• ii-agentIntelligent InternetOpen-source, self-sovereign intelligent assistant framework for building and deploying agents.
• EvoScientistEvoScientistSelf-evolving, human-on-the-loop AI scientist for “vibe research.”
• goclawnextlevelbuilderGoClaw is OpenClaw rebuilt in Go for deploying AI agent teams with multi-tenant isolation, 5-layer security, and native concurrency.
• embabel-agentEmbabelAgent framework for the JVM; Embabel Agents combine domain logic, AI capabilities, and tool usage to achieve user goals.
• Command-codeCommand CodeAI coding agent that learns coding taste and supports custom agents, memory, skills, MCP servers, and plugins.
• evolverEvoMapOpen-source Evolver client for EvoMap: a GEP-powered self-evolving engine for AI agents with auditable evolution using Genes, Capsules, and Events.
• hermes-web-uiEKKOLearnAIOpen-source, self-hosted AI agent dashboard for Hermes Agent with streaming chat, multi-model routing, Kanban boards, usage analytics, and a web terminal.
• open-multi-agentopen-multi-agentTypeScript-native multi-agent orchestration that turns a goal into a task DAG automatically.
• agentopsAgentOpsDeveloper platform and Python SDK for monitoring, testing, debugging, benchmarking, and deploying AI agents and LLM apps.
• rllmrLLMrLLM is a project for training AI agents with reinforcement learning for LLM-based workflows.
• holaOSholaboss-aiholaOS is a local-first desktop app and “super agent for work” for long-horizon AI work.
• SuperAGITransformerOptimusSuperAGI is a dev-first open-source autonomous AI agent framework for building, managing, and running useful autonomous agents.
• browser-harnessBrowser UseOpen-source, self-healing CDP-based harness for browser agents that helps LLMs complete browser tasks.
• eigentEigent AIOpen-source cowork desktop for multi-agent workflows and a local, free alternative to Claude Cowork.
• langchain4jLangChain4jOpen-source Java library for building LLM-powered applications on the JVM.
• GenericAgentlsdefineGenericAgent is a token-efficient, self-evolving LLM agent project from lsdefine.
• camelCAMEL-AIOpen-source CAMEL-AI framework for building multi-agent AI systems and studying the scaling laws of agents.
• leonLeon AIOpen-source personal assistant that can live on your server and supports text or voice interaction, including offline use for privacy.
• RagaAI-CatalystRagaAIAutomated AI evaluation and Test-and-Fix platform for LLM, RAG, and agentic applications.
• openfangRightNow-AIOpen-source Agent Operating System from RightNow AI.
• pydantic-aiPydanticAI Agent Framework, the Pydantic way.
• ppt-masterHugo He / hugohe3Open-source project that uses AI to generate real, editable PowerPoint PPTX files from documents.
• haystackdeepset-aiOpen-source AI orchestration framework for building context-engineered, production-ready LLM applications.
• AionUiiOfficeAIFree, local, open-source Cowork app for running and customizing AI agents on your computer.
• openai-agents-pythonOpenAIOfficial OpenAI Python Agents SDK: a lightweight framework for building agentic AI apps and multi-agent workflows.
• swarmOpenAIEducational framework exploring ergonomic, lightweight multi-agent orchestration from OpenAI.
• Open-AutoGLMzai-orgOpen phone agent model and framework for using AI to understand smartphone screens and execute automated phone operations.
• nocobaseNocoBaseOpen-source AI + no-code platform for building business systems with WYSIWYG editing and AI coding-agent collaboration.
• mastraMastraOpen-source TypeScript framework and platform for building AI-powered applications and agents.
• OpenCLIjackwenerMake any website into a CLI and let AI agents use your logged-in browser.
• activepiecesActivepiecesOpen-source AI-first automation platform for teams, with native AI pieces, agents, an agent library, and MCP support.
• CopilotKitCopilotKitFrontend stack for agents and generative UI for React and Angular applications.
• cherry-studioCherryHQAI productivity studio with smart chat, autonomous agents, 300+ assistants, and unified access to frontier LLMs.
• crewAICrewAIFramework for orchestrating role-playing, autonomous AI agents that work together on complex tasks.
• rufloruvnetAgent orchestration platform for Claude focused on multi-agent swarms, autonomous workflows, and conversational AI systems.
• BettaFish666ghjBettaFish is a multi-agent public opinion analysis assistant for analyzing social platforms and public comments.
• daily_stock_analysisZhuLinsenLLM-powered stock analysis system for A/H/US markets with market data, real-time news, decision dashboards, and multi-channel notifications.
• learn-claude-codeshareAI-labA nano Claude Code-like agent harness project for learning harness engineering for real agents.
• superpowersobraAn agentic skills framework and software development methodology.
• hermes-agentNous ResearchOpen-source MIT-licensed autonomous agent from Nous Research that lives on your server, remembers what it learns, and gets more capable over time.