awesome-ai-agents-2026

Contents

• AI Registry Explorer
• Quick Start
• Submit an Agent
• Orchestration Frameworks
• Coding Agents
• Memory and Context
• Multi-Agent Systems
• Agent Communication Protocols
• MCP (Model Context Protocol)
• Browser and Computer Use Agents
• Agent Tooling and Infrastructure
• Low and No-Code Builders
• Voice and Multimodal Agents
• Safety Guardrails and Observability
• Agent Interfaces and UIs
• Agent Deployment and Hosting
• Agent Evaluation and Benchmarks
• Agent Testing & Debugging
• Industry-Specific Agents
• Finance
• Healthcare
• Legal
• Manufacturing & Industrial
• Retail
• Government & Compliance
• Learning Resources
• Agent Communication
• Data Pipeline and Workflow
• Secure Execution Environments
• Creative AI Agents
• Image Generation
• Video Generation
• Music and Audio Generation
• 3D and Design Generation
• Customer Support and CRM Agents
• Support Automation
• AI-Powered CRM Platforms
• Sales and Outreach Agents
• Voice Agent Platforms
• Deep Research Agents
• Prompt-to-App Builders
• Multi-Agent Consumer Platforms
• Open-Source Models for Agents
• Market Stats and Benchmarks 2026
• Market Size
• Adoption Rates
• Top Use Cases
• Fastest Growing Open-Source Projects
• Notable Valuations
• Local and Self-Hosted AI
• AI Governance and Compliance
• Cybersecurity Agents
• Healthcare and Therapy Agents
• Translation and Localization Agents
• Newsletters and Communities
• Changelog
• Contributing
• Star History

---

AI Registry Explorer

Interactive Next.js 14 web application for searching, filtering, and exploring AI agents, frameworks, and infrastructure tools.

Quick Start

npm install && npm run build:data && npm run dev

Live Web Registry: https://ai-agent-registry-kappa.vercel.app
Local Web App: http://localhost:3000
Docker: docker compose up --build
One-Click Deploy:

Submit an Agent

Have an AI agent, framework, tool, skill, or infrastructure project worth sharing?

• Web Submission Portal: Submit a Resource
• GitHub Issue Form: Open No-Code Submission Issue

---

Orchestration Frameworks

• Agency Swarm 🚀 [Python] [Multi-Agent] - Orchestrates multi-agent systems built on the OpenAI Assistants API with role-based collaboration.
• AgentDescent 🔬 [Python] [Research] - Evolves agent skills, prompts, and harness modules by merging diffs from parallel workers against held-out reward.
• AgentScope 🚀 [Python] [Multi-Agent] - Alibaba multi-agent framework with distributed deployment and fault tolerance for production use.
• Agno 🌱 [Python] [Multi-Agent] - Multi-agent framework with a runtime and control plane for managing agent deployments at scale.
• Anthropic SDK 🚀 [Python] [Anthropic] - Official Claude SDK with tool use, computer control, and streaming for building Claude-native agents.
• AutoGen 🚀 [Python] [Multi-Agent] - Event-driven multi-agent framework merged with Semantic Kernel for production workflows.
• AWS Bedrock Agents 🚀 [Python] [Compliance] - Managed AWS infrastructure for agents with Bedrock models, enterprise compliance, and auto-scaling deployment.
• CAMEL 🌱 [Python] [Multi-Agent] - Role-based multi-agent simulation framework for collaborative reasoning and task completion.
• CopilotKit 🌱 [TypeScript] [Multi-Agent] - Open-source platform for embedding custom AI copilots into React applications.
• CrewAI 🚀 [Python] [Multi-Agent] - Production multi-agent framework with 5K+ GitHub stars, role-based collaboration, async execution, and 1500+ company adoptions.
• CUGA 🌱 [Python] [MCP] - Generalist agent harness from IBM Research that runs tasks across OpenAPI, MCP, and LangChain tools and the web, with policy guardrails and human approval.
• Doforu 🔬 [Desktop] [Multi-Agent] - Orchestrates parallel sub-agents on your desktop and verifies their results before merging.
• DSPy 🌱 [Python] [Research] - Stanford framework that programmatically optimizes LLM prompts using machine learning.
• Ontheia 🌱 [TypeScript] [MCP] - Self-hosted MCP-native agent platform with visual workflow automation and multi-provider support.
• Google ADK 🌱 [Python] [Multi-Agent] - Modular agent dev kit with native Gemini and Vertex AI integration.
• Haystack 🚀 [Python] [RAG] - Production-ready AI orchestration framework focused on building customizable LLM applications and RAG pipelines.
• LangGraph 🚀 [Python] [LangChain] - Stateful, graph-based agent workflow framework from LangChain.
• Letta 🌱 [Python] [Memory] - Stateful agents with built-in long-term memory and a REST API server.
• LightAgent 🌱 [Python] [MCP] - Lightweight agent framework with memory, MCP/SSE, Tree-of-Thought planning, and LightSwarm collaboration.
• LlamaIndex 🌱 [Python] [RAG] - The framework for connecting LLMs to your data, with indexing and retrieval capabilities.
• LangSmith Fleet 🚀 [Cloud] [LangChain] - No-code agent orchestration platform with visual workflow builder and enterprise deployment.
• Mastra 🌱 [TypeScript] [MCP] - Opinionated TypeScript framework with RAG, observability, and MCP support built in.
• MicroAgent 🌱 [TypeScript] [Multi-Agent] - Lightweight agent framework with self-editing prompts and code for minimal-overhead agent loops.
• Modus 🔬 [WebAssembly] [Serverless] - Serverless framework for high-throughput agent workloads with minimal cold starts.
• Open-AutoGLM 🔬 [Python] [Mobile] - Open-source phone agent framework for building mobile device automation agents.
• OpenAI Agents SDK 🚀 [Python] [Multi-Agent] - Lightweight multi-agent SDK with tracing and guardrails from OpenAI.
• PraisonAI 🚀 [Python] [MCP] - Production multi-agent framework with self-reflection, MCP integration, and workflow automation.
• PydanticAI 🌱 [Python] [Pydantic] - Type-safe agent framework from the Pydantic team with a FastAPI-style developer experience.
• Semantic Kernel 🚀 [C#] [Microsoft] - Microsoft enterprise SDK for Python, C#, and Java with modular plugins, memory, and goal planning.
• Smolagents 🌱 [Python] [Research] - Hugging Face code-first framework where agents write and execute Python instead of JSON tool calls.
• Strands Agents SDK 🌱 [Python] [AWS] - AWS model-driven agent SDK with native Bedrock integration.
• Upsonic 🌱 [Python] [MCP] - Minimalist agent framework with MCP support and rapid setup for quick prototyping.

Coding Agents

• agent-manager 🌱 [Go] [Multi-Agent] - Runs coding-agent CLIs side by side in persistent tmux sessions with live status, worktrees, and diff review.

• Aider 🌱 [Python] [CLI] - Terminal-first pair programmer that edits code in local repos, preserves Git history, and supports multi-file changes.

• Amazon Q Developer 🚀 [Python] [IDE] - AWS-native AI coding assistant with Lambda, CloudWatch, infrastructure support, and security scanning.

• Atomic Agent 🌱 [TypeScript] [Local] - Local-first CLI and TUI coding agent running open-weight models on your machine with no API key.

• AutoGPT 🌱 [Python] [CLI] - Mature autonomous agent platform with Forge framework and public benchmarks for evaluating agent capabilities.

• Claude Code 🚀 [TypeScript] [Anthropic] - Terminal-first agentic coding from Anthropic with Computer Use integration, multi-file edits, persistent shell sessions, Git operations, and fine-tuning support.

• Cline 🌱 [TypeScript] [VS Code] - Autonomous coding agent in your IDE that creates/edits files, runs commands, and uses the browser with permission-gated steps.

• CodeRabbit 🚀 [Cloud] [GitHub] - AI-powered PR reviewer with inline suggestions, security scanning, and automatic code quality checks.

• Codex CLI 🚀 [Python] [OpenAI] - OpenAI’s lightweight, open-source terminal coding agent with fast execution and strong benchmark scores.

• Codex-CLI 🌱 [C#] [CLI] - CLI tool that turns natural language commands into Bash, ZShell, and PowerShell equivalents.

• Continue 🌱 [TypeScript] [VS Code] - Source-controlled AI checks enforceable in CI, powered by the open-source Continue CLI.

• Cursor 🚀 [TypeScript] [IDE] - AI-native IDE with parallel Cloud Agents and 100K+ enterprise users.

• Devika 🌱 [Python] [Docker] - Open-source AI software engineer that understands high-level instructions and writes full-featured code autonomously.

• Devin 🚀 [Cloud] [GitHub] - Fully autonomous AI software engineer showing 8-12x efficiency gains in production (Nubank case study), cloud sandbox with persistent state, GitHub integration, and enterprise adoption growing 40% MoM.

• fractal 🌱 [Python] [Multi-Agent] - Orchestrates Claude Code, Codex, and other agents in per-node Git worktrees with recursive delegation and a live TUI.

• Frink 🔬 [Desktop] [Local] - Desktop app for running parallel coding agents and multi-step workflows triggered by schedules, webhooks, or events from tools like Linear and Sentry.

• gemini-cli 🌱 [Go] [CLI] - Open-source CLI agent bringing Gemini AI directly into your terminal.

• GitHub Copilot 🌱 [TypeScript] [VS Code] - Agent Mode in VS Code with Copilot Workspace for issue-to-PR workflows across Claude, GPT, and Gemini.

• Goose 🌱 [Rust] [CLI] - Extensible open-source CLI agent that installs, executes, edits, and tests with any LLM.

• JetBrains AI 🌱 [Kotlin] [JetBrains] - Deep AI integration across all JetBrains IDEs with context-aware completions and refactoring.

• Juggler 🌱 [Desktop] [Local] - Multi-client desktop/remote GUI agent with inspectable tool calls, branching-thread editable context, and plugin extensibility.

• Kiro 🚀 [Cloud] [IDE] - Spec-driven development agent that writes specs, auto-generates tasks, implements code, and automates DevOps workflows.

• Kolega Code 🔬 [Python] [CLI] - Terminal coding agent where the model writes its own multi-agent workflows across 15+ model providers.

• mu 🔬 [TypeScript] [CLI] - Coding agent that sends routine decisions, such as which tool output enters context, to a small judge model.

• Open Interpreter 🌱 [Python] [CLI] - Execute code locally via natural-language model instructions with a ChatGPT-like interface.

• opencode 🌱 [TypeScript] [Desktop] - Open-source coding agent available as a desktop app with a visual interface.

• OpenHands 🌱 [Python] [Docker] - AI-driven development platform that writes, tests, and deploys code autonomously.

• Orbi 🌱 [Python] [Self-Hosted] - Turns labeled GitHub Issues into independently reviewed pull requests, then merges reviewed heads and tags releases.

• Ordewell 🔬 [TypeScript] [Multi-Agent] - Turns one goal into an ordered, editable plan of coding-agent tasks, each with its own runner, model, and mode, verified by a completion marker in the runner’s output.

• Ouroboros 🌱 [Python] [MCP] - Pins an acceptance spec before the run and verifies the result, hiding grading commands from the executing agent.

• PR-Agent 🚀 [Python] [GitHub] - Open-source AI PR reviewer that auto-describes, reviews, and improves pull requests.

• Prime Agent 🚀 [TypeScript] [CLI] - Open-source RLM coding and research agent designed for long-running autonomous tasks.

• Qodo 🚀 [Cloud] [Security] - AI code review platform with context-aware PR validation and security analysis.

• Relay 🔬 [TypeScript] [CLI] - Runs a self-improving coding agent locally with your own models and re-runs tests before opening pull requests.

• RooCode 🌱 [TypeScript] [VS Code] - Cline fork with structured modes and reduced hallucinations for more reliable code generation.

• Snyk Code 🌱 [Cloud] [Security] - AI-powered security scanner with real-time vulnerability detection in agent-generated code.

• Sourcegraph Cody 🚀 [TypeScript] [IDE] - AI coding assistant that excels at large codebases with an enterprise context engine.

• SWE-agent 🚀 [Python] [GitHub] - Takes a GitHub issue and tries to automatically fix it.

• Tabnine 🚀 [TypeScript] [IDE] - Privacy-first AI code completion with on-premise deployment and codebase fine-tuning options.

• TaskWeaver 🚀 [Python] [Microsoft] - A code-first agent framework from Microsoft for planning and executing data analytics tasks.

• Windsurf 🌱 [TypeScript] [IDE] - AI-native IDE with Cascade agent for multi-step autonomous tasks and team workflows.

• workkit 🔬 [JavaScript] [GitHub] - Claude Code plugin that runs GitHub Issues as an agent pipeline, with a manager agent handing each issue to spec, build and verifier subagents.

• YYLO 🔬 [Python] [Multi-Agent] - Command-line orchestrator for coding agents with typed task, validation, merge, and release-readiness boundaries in a dedicated branch/worktree.

• Ivy Tendril - Open-source agentic software factory with an amazing UI that handles parallel Git worktrees for you, complete with programmatic verifications and fast review loops.

Memory and Context

• Acontext 🌱 [Python] [Local] - Manages agent skills and long-term memory as a layered data structure for persistent context.
• Agentage Memory 🌱 [Cloud] [MCP] - Cross-vendor shared memory layer exposed as a remote MCP server at memory.agentage.io/mcp (Streamable HTTP, OAuth 2.1 + PKCE + DCR) that Claude, Cursor, and ChatGPT read and write as plain markdown you own.
• Chroma 🌱 [Python] [Vector DB] - Lightweight, embeddable vector store for building memory-augmented AI agents with fast semantic retrieval.
• cognee 🌱 [Python] [Graph-Based] - Knowledge engine for AI agent memory, set up in 6 lines of code with graph-based knowledge extraction.
• Cortex Memory 🌱 [Python] [Vector DB] - Full-stack solution for agent memory covering extraction, vector search, and optimization.
• deja 🌱 [Go] [MCP] - Indexes the session transcripts 25 coding agents already write on the machine and serves them to any of the others, including sessions from before install; lexical retrieval, no LLM, no embeddings.
• Engrava 🔬 [Python] [Graph-Based] - Stores agent memory as a typed knowledge graph with hybrid search and a tamper-evident journal, embedded in SQLite.
• Enhanciar 🔬 [Cloud] [MCP] - A company brain that answers from your Slack, Gmail, Notion and Drive with the source, then drafts the reply or ticket for you to approve.
• Folio 🔬 [TypeScript] [MCP] - Git-backed team wiki that gives agents 40 MCP tools to search, read and edit pages and tables.
• graphiti 🌱 [Python] [Multi-Agent] - Build real-time knowledge graphs for AI agents with automatic entity extraction and linking.
• Kage 🌱 [TypeScript] [MCP] - Git-native memory for coding agents that stores decisions and fixes as repo files and verifies them against the codebase, withholding stale knowledge.
• LanceDB 🌱 [Rust] [Vector DB] - Serverless vector search database embedded directly in the agent process with no infrastructure needed.
• Langmem 🌱 [Python] [LangChain] - Helps agents learn and adapt from their interactions over time with persistent memory.
• Lians 🔬 [Python] [MCP] - Gives any AI agent local-first memory with corrections, point-in-time recall, and inspectable history.
• LWC 🔬 [Rust] [MCP] - Stores durable, source-grounded project memory and exposes bounded read-only retrieval to coding agents through MCP.
• Mem0 🌱 [Python] [Vector DB] - Memory layer for AI applications with long-term, short-term, and semantic memory extraction.
• Remio 🔬 [Desktop] [Memory] - Local-first AI memory desktop app that parses files into searchable vector indexes.
• Memoir 🔬 [Python] [Memory] - Git-like versioned semantic memory for AI agents with branching and commits.
• Memvid 🌱 [Python] [RAG] - Replace complex RAG pipelines with a serverless, single-file memory layer for instant retrieval.
• Milvus 🌱 [Go] [Vector DB] - Scales vector search to billions of embeddings for large-scale agent knowledge bases.
• Mnemoverse 🔬 [TypeScript] [MCP] - Hosted memory shared across Claude Code, Cursor and VS Code over MCP, with feedback that re-ranks what recall returns.
• Mori (森) 🌱 [Python] [MCP] - Sovereign shared memory layer for AI coding agents with zero-instrumentation capture via lifecycle hooks, a dream pipeline that distills sessions into curated governed memories, and support for Claude Code, Cursor, Codex, and Antigravity.
• Motorhead 🌱 [Rust] [Multi-Agent] - Manages conversation context windows for agents with automatic background summarization.
• Open Index 🔬 [Python] [MCP] - Builds typed knowledge graphs with hybrid search and read/write MCP tools for domain-specific agents.
• Pathway 🌱 [Python] [RAG] - Live data RAG engine with real-time streaming for agents that need up-to-the-second knowledge.
• Pinecone 🚀 [Cloud] [Vector DB] - Managed vector database with agent namespaces for multi-tenant isolation, hybrid search (vector + keyword), serverless auto-scaling, and $11B valuation.
• Qdrant 🌱 [Rust] [Vector DB] - High-performance vector similarity search engine with rich payload filtering for agent memory.
• RAGFlow 🌱 [Python] [RAG] - Open-source RAG engine with agent capabilities and deep document understanding for knowledge bases.
• Screenpipe 🚀 [Desktop] [MCP] - Local screen-text/audio memory via MCP, source-available under the Screenpipe Commercial License; configured cloud AI, transcription, sync, integrations and connected AI/MCP clients can transmit context off-device.
• SimpleMem 🌱 [Python] [Multimodal] - Efficient lifelong memory for LLM agents supporting both text and multimodal inputs.
• since-cutoff 🔬 [Python] [MCP] - Lists pinned Python dependency APIs that changed after a coding model’s training cutoff and writes AGENTS.md notes.
• Superdense 🔬 [TypeScript] [Memory] - Gives coding agents persistent memory of what worked across sessions, tracked against real-world outcomes.
• Supermemory 🌱 [TypeScript] [Vector DB] - Extremely fast and scalable memory engine and API designed for the AI era.
• Tree Ring Memory 🔬 [Rust] [Memory] - Manages local agent memory with recall, forgetting, audit trails, and session consolidation.
• Vestige 🌱 [Rust] [MCP] - Provides local-first memory for coding agents with FSRS-6 retention, active forgetting, and correction tools.
• Weaviate 🌱 [Go] [Vector DB] - Stores and searches vector embeddings with hybrid keyword and semantic retrieval for agent knowledge.
• Zep 🌱 [Python] [Multi-Agent] - Enriches agent long-term memory with automatic summarization, entity extraction, and search.

Multi-Agent Systems

• 5dive 🔬 [Shell] [Multi-Agent] - An open-source team of AI coding agents you text like coworkers, each with its own job.
• AgentVerse 🌱 [Python] [Multi-Agent] - Framework for building custom multi-agent environments to accomplish collaborative tasks.
• Bunkhouse 🔬 [TypeScript] [Multi-Agent] - Multitenant AI employees with a company inbox, org chart, and governed procedures for main-street business.
• Concord MCP 🌱 [TypeScript] [MCP] - Local-first MCP server giving coding agents shared work-state, including presence, file claims, messaging, handoffs and review packets, so multiple agents coordinate on one repo.
• EvoAgentX 🌱 [Python] [Multi-Agent] - Evaluates and evolves agentic workflows over time using automatic optimization.
• Hivekeep 🔬 [TypeScript] [Multi-Agent] - Runs a team of specialized self-hosted agents that collaborate, share memory, and build their own tools.
• Hivemoot 🚀 [Python] [GitHub] - Autonomous agent teams that collaboratively build software on GitHub.
• Markus 🌱 [TypeScript] [Multi-Agent] - Open-source AI workforce platform for building and running teams of AI agents with roles, task delegation, persistent memory, and inter-agent communication.
• MetaGPT 🌱 [Python] [Multi-Agent] - Simulates a full software company workflow from requirements to PRs using role-playing agents.
• NarraNexus 🌱 [Python] [Multi-Agent] - Ready-to-run AI agent team workspace by NetMind.AI whose agents remember, collaborate, and use tools from day one.
• P2PCLAW 🌱 [Python] [RAG] - Decentralized scientific research network with 14+ autonomous agents featuring P2P mesh, IPFS storage, and Lean 4 formal verification, self-hosted with Docker Compose.
• Raven 🌱 [Python] [Multi-Agent] - Plans complex tasks as DAGs and orchestrates built-in and third-party agents over ACP, CLI, or OpenAI-compatible APIs.
• Sqwakie 🔬 [TypeScript] [Multi-Agent] - Lets a team’s Claude Code and Codex agents message each other peer-to-peer and work shared kanban boards.
• Swarm 🚀 [Python] [OpenAI] - Lightweight framework for agent handoffs, context variables, and function calling patterns from OpenAI.
• Swarms Framework 🚀 [Python] [Multi-Agent] - Multi-agent orchestration for production use cases with scalability and reliability at its core.
• Yao Agents 🌱 [Go] [MCP] - Complete local AI execution platform with 30+ Experts, autonomous Robot orchestration via 5-stage Pipeline, API-based Robot-to-Robot calling, MCP support, and multi-channel messaging.

Agent Communication Protocols

The protocol layer that enables agents to discover tools, communicate with each other, and interoperate across ecosystems.

Protocol	Purpose	Creator	Status
MCP	Agent-to-tool	Anthropic	Standard
A2A	Agent-to-agent	Google	Growing
ACP	Agent communication	IBM/BeeAI	Early

MCP (Model Context Protocol)

• A2A Protocol 🚀 [Python] [Multi-Agent] - Google’s open protocol enabling AI agents to communicate, collaborate, and delegate tasks across frameworks.
• Aident Loadout 🔬 [Cloud] [MCP] - Hosted MCP server and CLI giving coding agents shared app connections, 400+ Skills, a credential vault, cost preview and audit logs.
• Arcade AI 🌱 [Python] [Multi-Agent] - Tool-use platform with authentication, authorization, and logging for agent-tool interactions.
• Chord 🔬 [Cloud] [MCP] - Gives agents DEX swap quotes, unsigned swap transactions, candles and wallet PnL on seven EVM chains.
• Composio 🌱 [TypeScript] [Multi-Agent] - Integration platform with 250+ pre-built tool connectors for AI agents and LLMs.
• Datacircle 🔬 [Cloud] [MCP] - Query your favorite B2B data APIs through us. Same request, same price, no markup.
• Docker MCP 🌱 [Go] [MCP] - Docker’s MCP gateway CLI plugin for running MCP servers in isolated containers.
• Equibles 🔬 [Cloud] [MCP] - Serves US stock data to agents over MCP: SEC filings, financials, earnings transcripts, insider trades and 13F holdings.
• HCS Agent Protocol 🌱 [TypeScript] [IDE] - Hedera open standards for agent identity with trustless P2P communication and 187K+ verified agents.
• HIG Doctor 🌱 [TypeScript] [MCP] - Apple HIG audit CLI and MCP server exposing design-guideline lookup and project audits for coding agents across SwiftUI, UIKit, React, Next.js, Flutter, Compose, HTML, and CSS.
• Hyper 🌱 [Cloud] [MCP] - Open-source Agent Skills and a hosted MCP connecting agents to 200+ marketing integrations across paid ads, SEO, analytics, social, and image and video generation, with a human-approval gate on every action.
• Jev Agent 🔬 [Cloud] [MCP] - Hosts a Jev model API and Streamable HTTP MCP so agents get typed decisions with probabilities.
• LogNorm 🔬 [Cloud] [MCP] - Hosted MCP server that hands a site’s SEO and AI-visibility backlog to Claude Code, Codex and Cursor.
• MCP Registry 🌱 [Python] [Multi-Agent] - Official Model Context Protocol specification and server implementations for standardized tool access.
• mcp-nest 🌱 [Python] [MCP] - Unified Model Context Protocol (MCP) server for executing code and managing files.
• NotFair 🚀 [Cloud] [MCP] - Hosted Google Ads MCP server for diagnosing, optimizing, and executing campaign changes via the Google Ads API with a human-approval gate.
• NotFair Skills 🚀 [TypeScript] [MCP] - Open-source Claude Code skills for SEO, GEO, Google Ads, and Meta Ads, connecting to live data through Google Ads MCP, Meta Ads MCP, Google Search Console MCP, and Google Analytics (GA4) MCP.
• Scalekit 🔬 [Cloud] [MCP] - Hosted MCP gateway and SDKs giving agents per-user OAuth, scoped tool permissions and audit logs across 500+ connectors.
• Seonix SEO MCP 🔬 [TypeScript] [MCP] - Audits any website for technical SEO, GEO/AEO readiness and page speed, then proposes and dry-runs minimal fixes.
• TableJourney 🔬 [Cloud] [MCP] - Hosted MCP server giving agents verified restaurants, food festival dates and bookable tours in 224 cities, and paying agents half of the booking commission.
• Tapetide MCP 🔬 [TypeScript] [MCP] - Gives agents quotes, financials, shareholding, screener and FII/DII flow data for about 8,200 NSE and BSE stocks.
• Toolhouse 🌱 [Python] [Multi-Agent] - Cloud-hosted tool infrastructure for agents with optimized execution and low-latency access.
• Xquik 🔬 [Cloud] [MCP] - Hosted X Twitter data MCP for search, follower export, monitors, and confirmation-gated writes.
• Zapier MCP Server 🌱 [Cloud] [MCP] - Connect agents to 7,000+ app integrations via MCP, powered by Zapier’s automation platform.
• zero-api-key-web-search 🌱 [Python] [MCP] - Free web search toolkit for AI agents with no API keys, MCP server support.

Browser and Computer Use Agents

• Agent S2 🌱 [Python] [Vision] - Open-source GUI automation framework for building desktop and browser agents with vision.
• AgentQL 🌱 [Python] [Multi-Agent] - AI-powered web scraping and automation with a semantic query language for page elements.
• aiFetchly 🔬 [TypeScript] [RAG] - Open-source desktop agent automating lead generation, knowledge library RAG, outreach, and scheduled workflows.
• AIHawk 🚀 [Python] [MCP] - Drives a real Firefox from plain-English instructions, with sixteen MCP tools and a standalone web UI.
• Airtop 🚀 [Cloud] [Multi-Agent] - Enterprise-grade cloud browser infrastructure for AI agent automation at scale.
• Browser Use 🌱 [Python] [Multi-Agent] - Open-source framework to let LLMs navigate and interact with any website programmatically.
• Browserbase 🚀 [Cloud] [Multi-Agent] - Scalable headless browser infrastructure purpose-built for running AI agents in production.
• Dia Browser 🌱 [Cloud] [Multi-Agent] - AI-native browser from Atlassian and the Browser Company with built-in agent workflows.
• Fellou 🌱 [Cloud] [Multi-Agent] - Transparent agentic browser with visual workflow editing and persistent agentic memory.
• Genspark 🌱 [Cloud] [Local] - On-device AI browser with 169+ local models that runs without internet connection.
• Grok Computer 🌱 [Cloud] [Multi-Agent] - Upcoming xAI desktop agent with mouse control and application automation capabilities.
• LaVague 🌱 [Python] [Multi-Agent] - Large Action Model framework to turn natural language instructions into browser automation.
• Manus 🚀 [Cloud] [Multi-Agent] - Autonomous digital employee with Browser Operator extension acquired by Meta for web-based task execution.
• MultiOn 🚀 [Cloud] [Multi-Agent] - Reliable web automation API with CAPTCHA handling for building production browser agents.
• OpenAI Operator 🚀 [Cloud] [OpenAI] - ChatGPT autonomous web agent with human checkpoints using Computer Use Agent technology.
• OpenAmer 🌱 [Python] [Local] - Open-source Windows desktop agent with native computer-use over the Chrome DevTools Protocol, in-process cognition tools, and peer-to-peer coordination between instances.
• OpenClaw 🌱 [Python] [Framework] - Autonomous open-source computer use and browser automation agent framework.
• Playwright MCP 🌱 [TypeScript] [MCP] - MCP server wrapping Playwright for seamless browser automation integration with LLM agents.
• Skyvern 🚀 [Python] [GitHub] - Production browser automation with 21.5K GitHub stars, 10M+ executed workflows, vision-first automation, and sub-300ms response times.
• Stagehand 🌱 [TypeScript] [Multi-Agent] - AI web browsing framework built on Playwright with natural-language selectors and actions.
• TODO for AI 🔬 [TypeScript] [Local] - Task-list agent that runs on your own machine and completes tasks through your shell, files and logged-in browser, with any model.

Agent Tooling and Infrastructure

Sandboxes, web scrapers, browser automation, and networking layers that agents depend on.

• Agent Bounties 🔬 [Rust] [MCP] - Coordinates verifiable digital bounty workflows designed for agents to post, fund, claim, solve, verify, and earn.
• Agent Payment Node 🔬 [TypeScript] [MCP] - Local CLI and stdio MCP payment runtime for AI agents on macOS Apple Silicon, with foreground owner approval for signing.
• AgentDock 🚀 [Python] [Docker] - Framework for building and deploying production-ready AI agents with composable node architecture.
• agent-qa 🌱 [TypeScript] [Testing] - Runs natural-language web and mobile tests with persistent memory and UI-change adaptation.
• AgentSandboxes 🔬 [Cloud] [MCP] - A free index of cloud sandboxes for AI agents with source-linked specs, a REST API and an MCP endpoint.
• Anansi Haven 🔬 [Cloud] [MCP] - Gives agents a free hosted home base with persistent memory, client-side encrypted storage, a job board and an agent directory over MCP, A2A and HTTP.
• Arcmira 🔬 [Cloud] [MCP] - Search indexed YouTube transcripts for timestamped passages, speaker appearances, entity mentions, sponsors and recommendations.
• codex-profiles 🚀 [Python] [OpenAI] - Bash CLI for switching OpenAI Codex CLI and Desktop profiles with isolated CODEX_HOME directories.
• Cohesivity 🔬 [Cloud] [MCP] - Headless backend for AI agents with databases, hosting, storage, and 15+ services via API and MCP, no API keys or signups required to get started.
• CompozyOS 🚀 [Go] [Multi-Agent] - Runs agent CLIs as a team on loops and schedules, with shared memory, permissions and approvals in one self-hosted daemon.
• Crawl4AI 🌱 [Python] [Multi-Agent] - Extracts structured data from web pages using LLM-friendly output formats optimized for agent ingestion.
• Docling 🌱 [Python] [IDE] - Parses PDFs, DOCX, and slides into structured text with deep layout understanding for document agents.
• E2B 🌱 [TypeScript] [Multi-Agent] - Cloud sandboxes for AI agents to run code securely in isolated environments.
• Engram 🌱 [Python] [Multi-Agent] - Universal bridge for multi-protocol AI agent systems with automated semantic mapping.
• Firecrawl 🌱 [TypeScript] [Multi-Agent] - Web scraping API built for LLMs that converts websites to clean, structured markdown.
• HumanEndpoint 🔬 [TypeScript] [MCP] - Human execution infrastructure for agents to request reviewed work, settle fees via x402 USDC, track jobs, and retrieve verifiable evidence.
• Jina Reader 🌱 [TypeScript] [Multi-Agent] - Converts any URL to LLM-ready clean text via a simple API prefix for agent ingestion.
• LlamaParse 🌱 [Python] [RAG] - GenAI-native document parser designed to extract complex tables and layouts for RAG pipelines.
• Loadout 🔬 [TypeScript] [IDE] - Free version-pinned AI coding rule packs for Cursor, Claude Code, Copilot, and AGENTS.md so agents stop inventing outdated APIs.
• Marker 🌱 [Python] [CLI] - Converts PDF documents to markdown with high accuracy for tables, equations, and figures.
• Notte 🚀 [Python] [Pipeline] - Browser automation engine optimized for production AI pipelines.
• Pilot Protocol 🌱 [Go] [Multi-Agent] - Networking stack for distributed agent systems with encrypted tunnels.
• Playwright 🌱 [TypeScript] [Testing] - Automates Chromium, Firefox, and WebKit browsers with a single cross-language API for agent-driven testing.
• SandBase CLI 🔬 [TypeScript] [MCP] - Connects coding agents to 2,000+ AI models through one onboarding command.
• ScrapeGraphAI 🌱 [Python] [LangChain] - Python web-scraping library that uses LLMs to build intelligent scraping pipelines from natural-language instructions.
• Solaris Dev Shop 🔬 [Python] [Skill] - Free MIT-licensed skill library with a deterministic intake router and 61 copy-paste SKILL.md specialists.
• Speck 🔬 [TypeScript] [Stateful] - Moves persistent memory, evidence tracking, validation, and cognitive state into software for small local LLM agents.
• Superagent 🔬 [Desktop] [Local] - Gives Claude Code and Codex a real browser and iOS Simulator to drive locally on macOS.
• Surya 🌱 [Python] [CLI] - Runs OCR and layout detection on documents in 90+ languages for multilingual document agents.
• Tavily 🌱 [Python] [Multi-Agent] - Search API purpose-built for LLM agents providing real-time, accurate web data with source citations.
• traceAI 🌱 [Python] [Observability] - OpenTelemetry-native tracing for LLM and agent apps with 50+ framework integrations.
• Tracefold 🚀 [Rust] [MCP] - Reverse-execution gateway that escrows checked pre-commit inverses, enforces Cedar policy gates, and produces signed offline-verifiable DSSE receipts for tool mutations.
• Unstructured 🌱 [Python] [Pipeline] - Ingests and preprocesses documents across 25+ file types for downstream LLM and agent pipelines.
• Vectoree 🔬 [Cloud] [CLI] - A backend for coding agents with a model gateway, auth, Postgres and storage behind one key.

Low and No-Code Builders

• Activepieces 🌱 [TypeScript] [Docker] - Open-source Zapier alternative with AI-powered automation nodes and visual workflow builder.
• AgentGPT 🌱 [TypeScript] [Local] - Deploy AI agents in the browser with zero local setup required.
• Dify 🌱 [Python] [RAG] - Open-source LLM app development platform with visual workflow builder and RAG orchestration.
• FastAgency 🚀 [Python] [Multi-Agent] - Deploys multi-agent workflows as production APIs with minimal configuration and setup.
• Flowise 🌱 [TypeScript] [RAG] - Open-source drag-and-drop UI to build customized LLM flows, RAG pipelines, and agent systems.
• Langflow 🌱 [Python] [RAG] - Visual drag-and-drop builder for LLM workflows, RAG agents, and multi-step pipelines.
• Lindy 🌱 [Cloud] [No-Code] - No-code AI agent platform with 3000+ app integrations for business workflow automation.
• LlamaIndex Workflows 🌱 [Python] [RAG] - Event-driven orchestration framework for building complex agentic systems.
• LLMGraph 🔬 [Cloud] [RAG] - Visual canvas for building RAG chatbots and AI agents with one-click deploy to REST API or chat widget.
• Make 🌱 [Cloud] [RAG] - Visual workflow automation platform with AI capabilities and drag-and-drop scenario builder.
• n8n 🌱 [TypeScript] [Docker] - Open-source workflow automation with AI agent nodes combining visual and code-based orchestration.
• Relevance AI 🌱 [Cloud] [No-Code] - No-code AI agent builder for sales, support, and research use cases with team collaboration.
• Rivet 🌱 [TypeScript] [RAG] - Visual AI workflow builder with drag-and-drop interface for designing agent pipelines.
• Syl 🔬 [Cloud] [MCP] - Hosted app for building teams of AI teammates that share conversations, keep memory, and run scheduled routines.
• Temporal 🚀 [Go] [Multi-Agent] - Durable execution platform for long-running agent workflows with automatic retry and state persistence.
• Wordware 🌱 [TypeScript] [IDE] - Web-hosted IDE where domain experts collaborate with AI engineers to build agent workflows.
• Workato Agent Studio 🚀 [Cloud] [No-Code] - No-code platform for building, deploying, and governing enterprise AI agents that reason, orchestrate multi-step workflows, and take actions across business applications.
• Zapier AI 🌱 [Cloud] [No-Code] - Connects 7000+ apps with natural language workflow creation for no-code agent automation.

Voice and Multimodal Agents

• Agentset 🚀 [Python] [RAG] - Production RAG platform with reasoning, hybrid search, and full multimodal support.
• LiveKit Agents 🌱 [Python] [IDE] - Framework for building real-time, multimodal AI agents with voice, video, and data channels.
• Nuance AI 🚀 [Cloud] [CLI] - Enterprise speech and conversational AI platform for clinical and contact-center workflows with HIPAA-capable deployments.
• Google Cloud Speech-to-Text v2 🚀 [Cloud] [Pipeline] - Google Cloud streaming and batch speech recognition API v2 with improved accuracy, streaming, and noise suppression for real-time agent pipelines.
• Personal Jarvis 🔬 [Python] [Voice] - Voice-driven desktop assistant that takes mouse and keyboard and delegates heavy tasks to agent harnesses like Claude Code, Codex, and MCP.
• Pipecat 🚀 [Python] [Streaming] - Production-grade voice AI framework with sub-250ms latency, WebRTC support, multimodal (voice+vision+text), real-time streaming, and 70+ language support.
• qwen-audio-agent 🌱 [Desktop] [Voice] - Full-duplex voice runtime that drives coding agents (Claude Code, Codex, OpenCode, Kimi Code, and more) over ACP, keeping conversations going while background tasks run, with barge-in and a local wake word.
• Rasa 🌱 [Python] [Self-Hosted] - Open-source conversational AI framework with self-hosted NLU training and dialogue management.
• simulate-sdk 🌱 [Python] [Voice] - Persona- and scenario-driven SDK for simulating voice and text AI agents.
• Vapi 🌱 [Python] [Voice] - Platform for building voice AI agents with low-latency speech-to-speech capabilities.
• Vocode 🌱 [Python] [Streaming] - Open-source framework for building voice-based LLM agent applications with streaming support.
• Wispr Flow 🚀 [Cloud] [Voice] - Voice orchestration platform for multimodal AI agents with 50+ language support, workflow building, and enterprise integrations.

Safety Guardrails and Observability

• Agent OS 🌱 [Python] [Multi-Agent] - Kernel architecture for governing autonomous AI agents with policy enforcement.
• AgentDoG 🌱 [Python] [Multi-Agent] - Diagnostic guardrails that analyze full agent execution trajectories to detect instruction hijacking and tool misuse.
• AgentGuard 🌱 [Python] [Observability] - Runtime observability and guardrails for AI agents with loop detection and anomaly alerts.
• Bifrost 🌱 [Go] [MCP] - AI gateway that unifies model providers with automatic failover, load balancing, observability, and MCP support.
• agenttrace 🌱 [Go] [CLI] - Local-first TUI for AI coding agent session observability with tokens, cost, latency, tool failures, anomalies, reports, diffs, and CI health gates.
• APort Agent Guardrails 🌱 [Python] [Multi-Agent] - Pre-action authorization plugin for agent frameworks with policy-based access control.
• Patronus AI LYNX 🌱 [Cloud] [Testing] - Hallucination detection system beating GPT-4 baselines, with specialized testing for agent outputs and LLM-generated content quality.
• Arize Phoenix 🌱 [Python] [Observability] - Open-source observability platform built on OpenTelemetry for tracing, evaluating, and debugging AI agents.
• Braintrust 🌱 [TypeScript] [Evaluation] - Eval-driven development platform with experiment tracking and prompt optimization for agent quality.
• ClawMetry 🔬 [Python] [Observability] - Self-hosted observability and opt-in kill switch for coding agents, reading the session logs runtimes already write to disk (website).
• ElevenAgents 🚀 [Cloud] [Voice] - Voice agent platform from ElevenLabs for customer support automation with HIPAA compliance and multi-language support.
• DriftGuard 🌱 [Python] [Multi-Agent] - Semantic memory guardrails using causal graphs to prevent agents from repeating past failures.
• Galley 🔬 [Go] [Multi-Agent] - Pairs independently configured executors and supervisors with repository-defined quality gates and inspectable evidence for each coding attempt.
• Guardrails AI 🌱 [Python] [Multi-Agent] - Adds structural, type, and quality guarantees to LLM outputs for reliable agent responses.
• Helicone 🌱 [TypeScript] [Observability] - Open-source LLM observability with one-line integration for cost tracking and prompt analytics.
• Lakera Guard 🌱 [Cloud] [Security] - Real-time protection against prompt injection, data leakage, and toxicity in agent interactions.
• Laminar 🌱 [Rust] [Observability] - Open-source observability and analytics platform purpose-built for the full lifecycle of AI agents.
• Langfuse 🌱 [TypeScript] [Observability] - Open-source LLM observability platform for tracing, prompt versioning, and LLM-as-a-judge evaluations.
• LangSmith 🚀 [Cloud] [LangChain] - LangChain platform for tracing, testing, and evaluating agent performance with production monitoring.
• LLM Guard 🌱 [Python] [Security] - Security toolkit for scanning LLM inputs and outputs to prevent prompt injection and data leaks.
• Logfire 🌱 [Python] [Pydantic] - Python-native observability from the Pydantic team with deep integration for high-performance agent monitoring.
• MartinLoop 🔬 [TypeScript] [Compliance] - Governs autonomous coding agents, turning open-ended runs into budgeted, verified software work with signed outcome receipts.
• NeMo Guardrails 🌱 [Python] [Multi-Agent] - NVIDIA programmable guardrails toolkit for controlling and securing LLM-powered agent conversations.
• Noveum 🔬 [Cloud] [Observability] - Traces AI agent runs, turns real interactions into evaluation datasets, and compares model quality, latency, and cost.
• OrcaReplay 🔬 [TypeScript] [Observability] - Records a coding-agent run below the harness, then replays it offline or forks it onto another model; its MCP server lets an agent read and re-run its own past runs.
• Orchard Kit 🌱 [Python] [Security] - Modules for agent runtime security, self-audit trails, and collective cognition patterns.
• OWASP Top 10 for Agentic Apps 🌱 [Python] [Security] - Security framework covering goal hijacking, tool misuse, and cascading failure mitigations for agents.
• Pluribus 🔬 [TypeScript] [Observability] - Generates cross-tool agent context and privacy-safe evidence receipts for loaded authority, handoffs, and skill use.
• Rebuff 🌱 [Python] [Security] - Self-hardening prompt injection detection system for securing agent inputs against adversarial attacks.
• Agent Learning Kit 🌱 [Python] [Evaluation] - LLM evaluation framework with 50+ metrics, LLM-as-Judge, and guardrail scanners (jailbreak, PII, injection).
• Shipmoor 🔬 [Python] [Testing] - Local, deterministic verification layer for AI agent code: scans, test evidence, and a binding merge verdict without uploading source.
• SourceryKit 🔬 [Python] [Security] - Verifies an agent’s outbound requests and MCP handoffs against a source of truth using zero-knowledge proofs, logging each call and blocking anything off the trusted-endpoint allow-list.
• Future AGI 🌱 [Python] [Self-Hosted] - Self-hostable end-to-end agent engineering platform with tracing, evals, guardrails, and gateway.
• sofagent 🔬 [TypeScript] [Security] - Commit-time audit harness that blocks credential leaks and out-of-scope file changes before they land, with 28 rules and HMAC-chained audit logs.
• SUNGLASSES 🔬 [Python] [MCP] - Scans text and files locally for prompt injection, credential leaks and data exfiltration with 1,554 patterns.

Agent Interfaces and UIs

Frontend workspaces and chat interfaces with built-in agent plugins and tool-use capabilities.

• AionUi 🚀 [Desktop] [Multi-Agent] - Connects 20+ AI CLIs and any API key in a local, open-source multi-agent desktop workspace.
• AnythingLLM 🌱 [TypeScript] [RAG] - All-in-one AI application with RAG, agents, and multi-model support for desktop and Docker.
• DB-GPT 🌱 [Python] [Database] - Data interaction platform with local LLM support for 100% private database and analytics agents.
• LibreChat 🌱 [TypeScript] [IDE] - Self-hosted multi-model chat interface supporting all major AI providers with access control.
• LobeHub 🌱 [TypeScript] [Multi-Agent] - Modern platform for hybrid work and AI-driven collaboration with extensible agent teams and rapid integration.
• LobeChat 🌱 [TypeScript] [Multi-Agent] - Modern, open-source AI chat framework with a massive plugin ecosystem for autonomous agent capabilities.
• OpenHuman 🚀 [Rust] [Memory] - Self-hosted local-first personal AI assistant with a Rust core, desktop apps, knowledge-graph memory, skills, voice, and multi-channel messaging.
• Orkas 🔬 [Desktop] [Multi-Agent] - Runs parallel AI agents in a local-first desktop workspace with shared files, BYOK providers, and optional sync.
• OpenWebUI 🌱 [TypeScript] [RAG] - Extensible local AI interface with built-in RAG, tool use, and support for multi-agent workflows.
• FutureOS 🔬 [Rust] [CLI] - One approval-gated AI agent spanning terminal, desktop, mobile, and chat clients on a shared local Rust backend.
• lucinate 🌱 [Go] [TUI] - Multi-backend terminal AI chat client for OpenClaw, Hermes, Ollama, and OpenAI-compatible APIs with routines, multi-agent switching, and local agent skills.
• Tale 🔬 [TypeScript] [Multi-Agent] - Self-hostable platform for coordinating people and AI agents through delegated project tasks, persistent sandbox workspaces, and human-reviewed results.
• TecAdRise Bot 🔬 [TypeScript] [Desktop] - Open-source Windows Electron multi-agent desktop chat shell powered by the Cursor SDK.
• Vicoa 🔬 [Desktop] [Multi-Agent] - Agentic IDE orchestrating a team of coding agents from desktop, web, or mobile with real-time sync.

Agent Deployment and Hosting

• Kortix 🚀 [Self-hosted] [Agent Platform] - Open-source AI Operating System: agents, skills, company memory and 3,000+ connectors in one git repo you own; every session runs on an isolated machine and lands its work as a change request.

• AWS Bedrock AgentCore 🚀 [TypeScript] [Compliance] - Managed AWS infrastructure for Bedrock-based agents with compliance, scaling, and monitoring built in.

• Cerebras Inference 🌱 [Cloud] [Multi-Agent] - Fastest LLM inference delivering 1000+ tokens per second on Llama 3.3 70B with a free tier.

• Fireworks AI 🌱 [Cloud] [RAG] - Serverless LLM inference with fine-tuning, RAG support, and free credits for rapid prototyping.

• Groq Cloud 🌱 [Cloud] [Multi-Agent] - Ultra-fast LPU-based LLM inference for Mixtral, Llama, and Gemma with a free API tier.

• Hivemeld 🚀 [Cloud] [Multi-Agent] - Hosted platform that deploys and runs teams of autonomous AI agents on a company’s real backlog across engineering, growth, and support in a shared workspace.

• Modal 🌱 [Python] [Serverless] - Serverless GPU compute purpose-built for AI workloads with fast cold starts and Python-native deployment.

• Northflank 🌱 [Cloud] [Kubernetes] - Full-stack platform with GPU orchestration, Git-based CI/CD, and bring-your-own-cloud support.

• Railway 🚀 [Cloud] [Stateful] - One-click deploy from GitHub with persistent volumes and databases for stateful agent deployments.

• Together AI 🌱 [Cloud] [Multi-Agent] - Inference API hosting 200+ open models with fast generation and a free tier for developers.

• Trigger.dev 🌱 [TypeScript] [Serverless] - Background job platform with cron, webhook, and event triggers purpose-built for long-running agent tasks.

Agent Evaluation and Benchmarks

• AgentBench 🌱 [Python] [Benchmark] - Comprehensive benchmark for evaluating LLMs as agents across 8 distinct environments.
• ARC-AGI-2 🌱 [Python] [Benchmark] - Frontier benchmark for measuring general intelligence capabilities in AI agents beyond pattern matching.
• ClawBench 🔬 [Python] [Benchmark] - Evaluates web agents on 283 real-world tasks across 163 live websites with interception and trace-based scoring.
• GAIA Benchmark 🌱 [Python] [Benchmark] - Benchmark for General AI Assistants measuring real-world reasoning and tool use.
• Inspect AI 🌱 [Python] [Evaluation] - Framework for evaluating large language models with composable tasks and scoring.
• Personal Agent Bench 🔬 [Browser] [Evaluation] - Takes consumer personal AI agents apart hands-on and scores them on real errands, reporting each result with a confidence interval and no composite score.
• [SWE-bench](https://github.com/SWE-bench/SWE

README 内容过长，已截取前 64 KB。完整内容请前往 GitHub 仓库查看。