# AgentSpot — full listing index

> A curated, automatically refreshed directory where you can discover which AI agents, reusable skills, and agent tools are popular — ranked by real, transparent metrics. A hub, not a platform: browse and click out.
> Generated 2026-08-03. Data license: CC-BY-4.0 (https://creativecommons.org/licenses/by/4.0/). Attribution: AgentSpot (https://agentspot.dev).

## Agents

- **agno** (Apache-2.0) — A Python framework and platform for building, running, and managing AI agents, with built-in memory, tools, and multi-agent teams. [41,141 stars, 2,303,831 pypi downloads/30d] Page: https://agentspot.dev/agents/agno/ | Repo: https://github.com/agno-agi/agno | Site: https://docs.agno.com
- **anything-llm** (MIT) — An all-in-one, self-hostable desktop and web app for running local-first AI agents over your own documents, with RAG, tools, and multi-user support. [63,244 stars] Page: https://agentspot.dev/agents/anything-llm/ | Repo: https://github.com/Mintplex-Labs/anything-llm | Site: https://anythingllm.com
- **autogen** (CC-BY-4.0) — Microsoft's framework for building multi-agent AI applications, where several conversable agents coordinate, call tools, and hand off work to solve tasks together. [59,715 stars, 1,082,455 pypi downloads/30d] Page: https://agentspot.dev/agents/autogen/ | Repo: https://github.com/microsoft/autogen | Site: https://microsoft.github.io/autogen/
- **AutoGPT** (Other (see repo)) — One of the earliest autonomous GPT agents, now a platform for building, deploying, and running AI agents that chain tasks toward a goal with minimal human input. [185,512 stars] Page: https://agentspot.dev/agents/autogpt/ | Repo: https://github.com/Significant-Gravitas/AutoGPT | Site: https://agpt.co
- **browser-use** (MIT) — Lets AI agents control a real web browser to navigate sites and complete tasks, turning natural-language goals into automated clicks, form fills, and data extraction. [104,593 stars, 178,945,759 pypi downloads/30d] Page: https://agentspot.dev/agents/browser-use/ | Repo: https://github.com/browser-use/browser-use | Site: https://browser-use.com
- **crewAI** (MIT) — A framework for orchestrating teams of role-playing autonomous agents that collaborate on complex tasks, each with its own goal, tools, and backstory. [55,461 stars, 11,322,805 pypi downloads/30d] Page: https://agentspot.dev/agents/crewai/ | Repo: https://github.com/crewAIInc/crewAI | Site: https://crewai.com
- **cua** (MIT) — Open-source infrastructure for computer-use agents: sandboxes, SDKs, and benchmarks for AI that controls full macOS, Linux, and Windows desktops. [19,628 stars] Page: https://agentspot.dev/agents/cua/ | Repo: https://github.com/trycua/cua | Site: https://cua.ai
- **deer-flow** (MIT) — An open-source long-horizon agent harness from ByteDance that researches, writes code, and produces content using sandboxes, memory, tools, and subagents. [76,942 stars] Page: https://agentspot.dev/agents/deer-flow/ | Repo: https://github.com/bytedance/deer-flow | Site: https://deerflow.tech
- **gemini-cli** (Apache-2.0) — Google's open-source AI agent that brings Gemini into the terminal for coding, automation, and MCP-connected workflows. [105,970 stars] Page: https://agentspot.dev/agents/gemini-cli/ | Repo: https://github.com/google-gemini/gemini-cli | Site: https://geminicli.com
- **langchain** (MIT) — A widely adopted framework for building LLM applications and agents, with composable abstractions for prompts, memory, tool calling, and retrieval across many model providers. [141,697 stars, 324,514,125 pypi downloads/30d] Page: https://agentspot.dev/agents/langchain/ | Repo: https://github.com/langchain-ai/langchain | Site: https://docs.langchain.com/langchain/
- **langgraph** (MIT) — A library from the LangChain team for building stateful, multi-step agent workflows as graphs, with durable execution, branching, and human-in-the-loop control. [37,210 stars, 64,587,834 pypi downloads/30d] Page: https://agentspot.dev/agents/langgraph/ | Repo: https://github.com/langchain-ai/langgraph | Site: https://docs.langchain.com/oss/python/langgraph/
- **letta** (Apache-2.0) — A platform for building stateful agents with long-term memory that persists across sessions, letting agents learn and improve over time (formerly MemGPT). [23,776 stars, 162,786 pypi downloads/30d] Page: https://agentspot.dev/agents/letta/ | Repo: https://github.com/letta-ai/letta | Site: https://docs.letta.com/
- **mini-swe-agent** (MIT) — A deliberately minimal (~100-line) coding agent that resolves GitHub issues and assists at the command line, while scoring competitively on SWE-bench. [5,766 stars, 13,492,987 pypi downloads/30d] Page: https://agentspot.dev/agents/mini-swe-agent/ | Repo: https://github.com/SWE-agent/mini-swe-agent | Site: https://mini-swe-agent.com
- **pydantic-ai** (MIT) — An agent framework from the Pydantic team that brings type-safe, structured outputs and validation to LLM applications and tool calling. [18,497 stars, 20,128,114 pypi downloads/30d] Page: https://agentspot.dev/agents/pydantic-ai/ | Repo: https://github.com/pydantic/pydantic-ai | Site: https://pydantic.dev/pydantic-ai
- **SuperAGI** (MIT) — A developer-first open-source framework for building, running, and managing autonomous AI agents, with a GUI, tool integrations, and concurrent agent support. [17,619 stars] Page: https://agentspot.dev/agents/superagi/ | Repo: https://github.com/TransformerOptimus/SuperAGI | Site: https://superagi.com/

## Skills

- **agents** (MIT) — A large community collection of subagents and skills packaged as plugins for Claude Code, Codex CLI, Cursor, and other agent harnesses. [37,878 stars] Page: https://agentspot.dev/skills/agents/ | Repo: https://github.com/wshobson/agents | Site: https://sethhobson.com
- **awesome-claude-code** (Other (see repo)) — A curated collection of slash commands, workflows, CLAUDE.md files, and tooling for getting the most out of coding agents. [49,958 stars] Page: https://agentspot.dev/skills/awesome-claude-code/ | Repo: https://github.com/hesreallyhim/awesome-claude-code
- **awesome-claude-skills** (Other (see repo)) — A curated list of Claude Skills, resources, and tools for customizing AI agent workflows. [67,653 stars] Page: https://agentspot.dev/skills/awesome-claude-skills/ | Repo: https://github.com/ComposioHQ/awesome-claude-skills
- **claude-plugins-official** (Apache-2.0) — Anthropic's official, managed directory of high-quality Claude Code plugins, which bundle skills, commands, and MCP servers. [32,090 stars] Page: https://agentspot.dev/skills/claude-plugins-official/ | Repo: https://github.com/anthropics/claude-plugins-official | Site: https://code.claude.com/docs/en/plugins
- **generative-ai-for-beginners** (MIT) — Microsoft's 21-lesson course teaching the fundamentals of building generative-AI applications, from prompting to building agents. [112,962 stars] Page: https://agentspot.dev/skills/generative-ai-for-beginners/ | Repo: https://github.com/microsoft/generative-ai-for-beginners
- **goose-skills** (Other (see repo)) — A library of growth and go-to-market skills for coding agents, covering ads, SEO, content, and lead generation. [1,006 stars] Page: https://agentspot.dev/skills/goose-skills/ | Repo: https://github.com/gooseworks-ai/goose-skills | Site: https://gooseworks.ai/cli
- **playwright-skill** (MIT) — A model-invoked Claude Code skill for browser automation with Playwright, letting an agent write and run its own end-to-end tests. [2,905 stars] Page: https://agentspot.dev/skills/playwright-skill/ | Repo: https://github.com/lackeyjb/playwright-skill
- **Prompt-Engineering-Guide** (MIT) — A comprehensive open guide to prompt engineering, context engineering, RAG, and agents, with lessons, papers, and worked examples. [76,434 stars] Page: https://agentspot.dev/skills/prompt-engineering-guide/ | Repo: https://github.com/dair-ai/Prompt-Engineering-Guide | Site: https://www.promptingguide.ai/
- **prompts.chat** (Other (see repo)) — A community collection of reusable prompts (formerly Awesome ChatGPT Prompts) that you can share, discover, and self-host for your organization. [165,647 stars] Page: https://agentspot.dev/skills/prompts-chat/ | Repo: https://github.com/f/prompts.chat | Site: https://prompts.chat
- **skills** (Other (see repo)) — Anthropic's public repository of Agent Skills — packaged instructions and resources (the SKILL.md format) that teach agents to perform specific tasks reliably. [160,876 stars] Page: https://agentspot.dev/skills/skills/ | Repo: https://github.com/anthropics/skills

## Tools

- **chroma** (Apache-2.0) — An open-source search and retrieval engine (vector database) for building AI applications with embeddings-based memory and RAG. [28,781 stars, 12,782,396 pypi downloads/30d] Page: https://agentspot.dev/tools/chroma/ | Repo: https://github.com/chroma-core/chroma | Site: https://www.trychroma.com/
- **chrome-devtools-mcp** (Apache-2.0) — An official MCP server exposing Chrome DevTools to coding agents, letting them inspect, debug, and drive a real Chrome browser. [46,860 stars, 10,947,136 npm downloads/30d] Page: https://agentspot.dev/tools/chrome-devtools-mcp/ | Repo: https://github.com/ChromeDevTools/chrome-devtools-mcp | Site: https://npmjs.org/package/chrome-devtools-mcp
- **composio** (MIT) — A platform giving agents 1000+ tool integrations, tool search, managed authentication, and a sandboxed workbench for taking real-world actions. [29,221 stars, 1,705,069 pypi downloads/30d] Page: https://agentspot.dev/tools/composio/ | Repo: https://github.com/ComposioHQ/composio | Site: https://docs.composio.dev
- **context7** (MIT) — An MCP server that feeds up-to-date, version-specific library documentation into LLMs and AI code editors so generated code targets current APIs. [59,063 stars, 4,079,569 npm downloads/30d] Page: https://agentspot.dev/tools/context7/ | Repo: https://github.com/upstash/context7 | Site: https://context7.com
- **dify** (Other (see repo)) — A production-ready platform for building agentic LLM applications and workflows, combining a visual builder, RAG, and model orchestration. [148,720 stars] Page: https://agentspot.dev/tools/dify/ | Repo: https://github.com/langgenius/dify | Site: https://dify.ai
- **E2B** (Apache-2.0) — Open-source secure cloud sandboxes that let AI agents run code and use real tools in isolated environments. [12,963 stars, 4,909,725 pypi downloads/30d] Page: https://agentspot.dev/tools/e2b/ | Repo: https://github.com/e2b-dev/E2B | Site: https://e2b.dev/docs
- **exa-mcp-server** (MIT) — An MCP server exposing Exa's neural web search and crawling so agents can retrieve current information from the web. [4,704 stars, 95,944 npm downloads/30d] Page: https://agentspot.dev/tools/exa-mcp-server/ | Repo: https://github.com/exa-labs/exa-mcp-server | Site: https://exa.ai/docs/reference/exa-mcp
- **fastmcp** (Apache-2.0) — A Pythonic framework for building Model Context Protocol servers and clients quickly, handling the protocol plumbing for you. [26,183 stars, 97,342,888 pypi downloads/30d] Page: https://agentspot.dev/tools/fastmcp/ | Repo: https://github.com/PrefectHQ/fastmcp | Site: https://gofastmcp.com
- **Figma-Context-MCP** (MIT) — An MCP server that hands Figma layout and design data to AI coding agents so they can build UIs that match a design. [15,385 stars, 393,070 npm downloads/30d] Page: https://agentspot.dev/tools/figma-context-mcp/ | Repo: https://github.com/GLips/Figma-Context-MCP | Site: https://www.framelink.ai/
- **firecrawl** (AGPL-3.0) — An API that crawls, scrapes, and searches the web at scale and returns clean, LLM-ready markdown for agents and RAG pipelines. [150,444 stars, 7,325,491 pypi downloads/30d] Page: https://agentspot.dev/tools/firecrawl/ | Repo: https://github.com/firecrawl/firecrawl | Site: https://firecrawl.dev
- **firecrawl-mcp-server** (MIT) — The official Firecrawl MCP server, adding web scraping and search to MCP-compatible clients like Claude and Cursor. [6,938 stars, 536,555 npm downloads/30d] Page: https://agentspot.dev/tools/firecrawl-mcp-server/ | Repo: https://github.com/firecrawl/firecrawl-mcp-server | Site: https://firecrawl.dev
- **github-mcp-server** (MIT) — GitHub's official MCP server, letting agents work with repositories, issues, pull requests, and Actions through the GitHub API. [31,412 stars] Page: https://agentspot.dev/tools/github-mcp-server/ | Repo: https://github.com/github/github-mcp-server
- **mcp-toolbox** (Apache-2.0) — Google's open-source MCP server for databases, giving agents secure, pooled access to query many SQL and NoSQL data stores. [15,948 stars] Page: https://agentspot.dev/tools/mcp-toolbox/ | Repo: https://github.com/googleapis/mcp-toolbox | Site: https://mcp-toolbox.dev/documentation/introduction/
- **mem0** (Apache-2.0) — A universal memory layer for AI agents and assistants that stores, recalls, and manages long-term user and session memory. [60,758 stars, 3,088,468 pypi downloads/30d] Page: https://agentspot.dev/tools/mem0/ | Repo: https://github.com/mem0ai/mem0 | Site: https://mem0.ai
- **n8n** (Other (see repo)) — A fair-code workflow automation platform with native AI features and hundreds of integrations, letting you wire agents and tools into visual, self-hostable pipelines. [196,337 stars] Page: https://agentspot.dev/tools/n8n/ | Repo: https://github.com/n8n-io/n8n | Site: https://n8n.io
- **open-webui** (Other (see repo)) — A self-hosted, user-friendly web interface for local and API-based LLMs, with support for Ollama, OpenAI-compatible endpoints, RAG, and tools. [145,320 stars] Page: https://agentspot.dev/tools/open-webui/ | Repo: https://github.com/open-webui/open-webui | Site: https://openwebui.com
- **playwright-mcp** (Apache-2.0) — Microsoft's MCP server that gives agents structured, accessibility-tree-based control of a browser via Playwright, without relying on screenshots. [35,037 stars, 24,126,253 npm downloads/30d] Page: https://agentspot.dev/tools/playwright-mcp/ | Repo: https://github.com/microsoft/playwright-mcp | Site: https://www.npmjs.com/package/@playwright/mcp