Tools (106)
Standalone tools agents use: browsers, sandboxes, code execution, search and more.
Machine-readable: /api/tools.json.
- A-MEM (Agentic Memory)
Research memory library for LLM agents that writes Zettelkasten-style linked notes and evolves them, backed by ChromaDB.
- Acontext (deprecated)
Archived and no longer maintained (Oct 2026): open-source skill memory layer that distilled agent runs into editable SKILL.md files agents fetched on later runs.
- Agent Lightning
Microsoft's lightweight agentic RL framework that trains models through real agent harnesses unchanged, using a proxy gateway, verl/vLLM trainer and Kubernetes rollouts.
- agent-browser
Vercel Labs' browser automation CLI built for AI agents.
- Agentless
Software-repair baseline using localization, candidate patches and test-based validation rather than an autonomous tool-using agent loop.
- agentmemory
Persistent memory server for coding agents (Claude Code, Copilot CLI, Cursor, Gemini CLI, Codex CLI, OpenCode and any MCP client) with hybrid search.
- Anchor Browser
Cloud browser infrastructure for computer-use and browser agents; agents can obtain an API key programmatically.
- Arcade
Actions runtime for enterprise AI agents: agent authorization, agent-optimized tools (MCP servers) and governance, plus an open-source MCP framework.
- arcade-mcp
Arcade's open-source Python framework for building MCP servers and tools, with authorized tool calling (OAuth, token refresh) and tool-call evals.
- Auth0 for AI Agents
Auth0's identity toolkit for AI agents: user authentication, delegated authorization, human-in-the-loop (CIBA) and fine-grained authorization for RAG.
- Axolotl
Apache-2.0 post-training / fine-tuning framework for a wide range of LLMs, with CLI, YAML configs, Docker and distributed-training options.
- Bevy
Free, open-source, data-driven game engine in Rust built on an entity component system (ECS).
- Browser Use
Open-source library for agents that use the browser, plus a cloud offering of browser agents and managed browsers via SDK, REST or CDP.
- Browserbase
Hosted headless-browser infrastructure that gives agents access to the whole web.
- Browserless
Hosted or self-hosted (Docker) headless browsers for AI agents, reachable over MCP, Puppeteer or Playwright.
- Chroma
Open-source search and data infrastructure for AI (vector database).
- Claude Memory Tool
Client-side Claude API tool that lets Claude create, read, update and delete files in a /memories directory to persist knowledge across conversations.
- Cognee
Open-source AI memory platform for agents that turns documents and conversations into graph, vector and relational memory.
- Composio
Tool platform for agents: 1,000+ toolkits with tool search, delegated authentication and sandboxed execution.
- Crawl4AI
Open-source LLM-friendly web crawler and scraper that turns any website into clean Markdown; self-host or use the hosted cloud.
- Descope Agentic Identity Hub
Manage authentication, authorization and external credentials for AI agents and MCP servers with Descope.
- Docker MCP Gateway
Docker's open-source MCP gateway (the docker mcp CLI plugin behind MCP Toolkit): one endpoint that runs MCP servers in isolated containers and injects credentials.
- Docling
Open-source document parser that converts PDF, Office, HTML, audio, images and more into Markdown, JSON or DocTags for gen-AI pipelines; includes an MCP server.
- EverOS
EverMind's local-first memory runtime for agents: conversations, files and trajectories stored as Markdown, indexed in SQLite and LanceDB, shared across coding tools.
- FastMCP
The fast, Pythonic framework for building MCP servers, clients and interactive applications.
- GBrain
Garry Tan's open-source agent brain: shared memory with sources, corrections and a knowledge graph, served to Claude, Codex, OpenClaw, Hermes and others over MCP.
- gh skill (GitHub CLI) (beta)
GitHub CLI command group to search, preview, install, list, update and publish agent skills from GitHub repositories (preview).
- GitHub Agentic Workflows (gh-aw)
GitHub Actions-based agentic workflows with Copilot, Claude, Codex or Antigravity to triage issues, investigate failures and propose changes, with safety controls.
- Godot Engine
Free, open-source, multi-platform 2D and 3D game engine, scripted in GDScript or C#. Several community MCP servers let agents drive its editor.
- Graphiti
Zep's open-source framework for building real-time knowledge graphs for AI agents.
- Guidance
Microsoft Research library for constrained LLM generation with a Python guidance language, token healing and multi-backend support.
- Hindsight
Open-source agent memory system from Vectorize focused on agents that learn: retain, recall and reflect operations over memory banks, with an MCP server and a managed cloud.
- HOL Guard
Local-first security layer that reviews AI coding agents' shell commands, file access, package installs and MCP tool calls before they run.
- Honcho
Open-source memory infrastructure for stateful agents that models people, agents, groups and projects over time; managed or self-hosted.
- Hy Memory
Tencent Hunyuan's dual-system cognitive memory engine and Python SDK for agents, with chat, working and coding memory pipelines plus the MemHub app.
- HyperAgent
Hyperbrowser's AI browser-automation library: Playwright plus natural-language commands (page.ai, extract, executeTask), stealth mode and action caching.
- Hyperbrowser
Fast cloud browsers for AI agents and automation, with an official MCP server.
- Kernel
Browsers-as-a-service for web agents and automations: anti-bot handling, reusable sessions, fast spin-up and autoscaling.
- LanceDB
Developer-friendly, open-source embedded retrieval library and multimodal lakehouse for AI.
- LangMem
LangChain library that helps agents learn from interactions: memory tools, a background memory manager and prompt optimization, with native LangGraph store integration.
- Letta
Platform for stateful agents with memory, identity and the ability to learn over time (formerly MemGPT); terminal UI, App Server, desktop app and Agent SDK.
- Lightpanda
Headless browser built from scratch for AI and automation, designed for speed and a minimal memory footprint.
- LightRAG
Graph-plus-vector RAG framework from HKU with a server, web UI and Ollama-compatible API; pluggable storage backends.
- LiteLLM
MIT Python SDK and self-hosted AI gateway that calls 100+ LLM providers in the OpenAI format. Paid Enterprise is separate and quote-based.
- LLaMA-Factory
Apache-2.0 toolkit to fine-tune 100+ LLMs and VLMs via zero-code CLI and the LLaMA Board Gradio web UI.
- llama.cpp
MIT C/C++ library and server for local LLM inference, widely used as a local OpenAI-compatible backend for coding agents.
- llguidance
High-performance structured-output / constrained-decoding engine from the Guidance project (Rust core, Python bindings).
- LocalAI
MIT self-hosted OpenAI-compatible engine that runs LLM, vision, voice and image backends locally, with Docker images as the documented install.
- m3 Memory
Local-first, MCP-native memory layer for coding agents (Claude Code, Codex, Gemini CLI, Antigravity, OpenCode, OpenClaw) with hybrid search and chat capture.
- magentic
Python decorators that turn typed functions into LLM prompts, combining structured outputs, streaming and tool calls with ordinary Python control flow.
- MarkItDown
Microsoft's lightweight Python utility for converting PDF, Office, images, audio, HTML and more into structure-preserving Markdown for LLMs.
- Massvai
Hosted AI app builder that turns a prompt into an exportable full-stack Next.js app with live preview, Supabase, GitHub sync and Vercel deploy.
- MCP Go SDK
Official Go SDK for Model Context Protocol servers and clients, maintained in collaboration with Google.
- MCP Python SDK
Official Python SDK for building Model Context Protocol servers and clients.
- MCP TypeScript SDK
Official TypeScript SDK for building MCP servers and clients on Node.js, Bun and Deno; v2 ships as @modelcontextprotocol/server and @modelcontextprotocol/client.
- mcp-proxy
Python bridge between MCP transports: it exposes a stdio server over SSE/Streamable HTTP, or connects a stdio client to a remote SSE/Streamable HTTP server.
- mcp-remote
npx tool that lets an MCP client that only supports local stdio servers connect to a remote MCP server, including OAuth flows.
- mcp-use
Full-stack MCP framework (TypeScript and Python) for building MCP servers and MCP Apps with typed tools, React views, a built-in Inspector and agent clients.
- mcpo
Open WebUI's MCP-to-OpenAPI proxy that exposes any MCP server as a REST/OpenAPI HTTP server, so OpenAPI-speaking agents and tools can call it.
- Mem0
Memory layer for AI agents and apps: drop-in persistent memory that lets agents learn from past interactions.
- Memobase
Open-source long-term memory for LLM apps that builds a structured profile and event timeline per user.
- Memori
Open-source, LLM- and datastore-agnostic memory layer for agents from Memori Labs: Python/TypeScript SDKs that wrap your LLM client, bring-your-own DB or Memori Cloud, plus MCP.
- MemoryOS
Hierarchical short/mid/long-term memory system for personalized agents, as a PyPI package and an MCP server.
- MemOS
Open-source memory operating system for LLMs and agents with a unified API for adding, retrieving, editing and deleting graph-structured, multi-modal memory.
- memU
Agent-driven personal memory stored as a shared LLM wiki across sessions, agents and devices; distills reusable skills from Codex, Claude Code, Cursor and OpenClaw history.
- Milvus
High-performance, cloud-native open-source vector database built for scalable approximate-nearest-neighbor search.
- MIRIX
Open-source multi-agent memory system: six memory types (core, episodic, semantic, procedural, resource, knowledge vault) managed by dedicated agents, with optional screen capture.
- MLC LLM
Apache-2.0 compiler and deployment engine for running LLMs natively on NVIDIA, AMD, Apple, Intel, web, iOS and Android.
- MLX LM
MIT Python package for text generation and fine-tuning of large language models on Apple silicon with MLX.
- ms-SWIFT
ModelScope SWIFT toolkit for training, fine-tuning, deploying and evaluating 100+ LLMs and multimodal models (Apache-2.0).
- Notte
Cloud browser infrastructure and web automation platform for AI and coding agents.
- Ollama
MIT app and API for running open models locally, plus a paid Ollama Cloud with usage credits and per-token prices.
- OpenRLHF
Apache-2.0 high-performance RLHF / post-training framework built on Ray, with support for multiple reward and policy-optimization recipes.
- Oumi
Apache-2.0 end-to-end foundation-model platform for training, evaluation, inference and agentic/tool-use post-training, installable as PyPI oumi 0.9.
- Outlines
Structured LLM output library from .txt (dottxt): JSON/schema/grammar constraints with broad model integrations.
- PEFT
Hugging Face Parameter-Efficient Fine-Tuning library (LoRA and related methods) for adapting large models without full fine-tuning.
- pgvector
Open-source vector similarity search extension for Postgres.
- Pinecone
Managed vector database for AI agents and applications: semantic search, knowledge retrieval and long-term memory at scale.
- Playwright
Microsoft's framework for web testing and automation across Chromium, Firefox and WebKit with a single API.
- Playwright CLI
CLI for common Playwright actions (record, inspect selectors, screenshots), recommended with skills as a token-efficient alternative to Playwright MCP for coding agents.
- Puppeteer
JavaScript API for controlling Chrome and Firefox, commonly used for headless browser automation.
- Qdrant
Open-source vector search engine and database written in Rust, self-hosted or as Qdrant Cloud.
- Redis Agent Memory
Redis's managed memory layer for agents (part of Redis Iris): session memory with TTL plus long-term memory extracted in the background, via REST, MCP and SDKs.
- SGLang
Apache-2.0 inference framework for LLM, vision-language and diffusion models, optimized for agentic workloads and large-scale serving.
- Skyvern
Automates browser-based workflows with AI: describe tasks in plain English and Skyvern drives a real browser using vision.
- Stagehand
Browserbase's SDK for browser agents to extract data and interact with any site: Playwright-style API and self-healing selectors in TypeScript, Python and Go.
- Steel
Open-source browser API and sandbox for AI agents, with managed cloud browsers offering stealth, proxies, CAPTCHA solving and session replays.
- Stytch Connected Apps (MCP auth)
Stytch Connected Apps: authorization for remote MCP servers and connected agent integrations.
- Supergateway
Runs MCP stdio servers over Streamable HTTP or SSE, or bridges a remote SSE/Streamable HTTP server back to stdio, using a single npx command.
- Supermemory
Memory and context engine for AI agents with a Memory API; can run fully locally or hosted.
- SurrealDB Agent Memory
Managed memory and knowledge layer for agents on SurrealDB: facts with provenance and supersession, hybrid retrieval, and a built-in MCP server.
- torchtune (deprecated)
PyTorch-native BSD-3-Clause library for authoring, fine-tuning and evaluating LLMs with native PyTorch training primitives.
- TRL
Hugging Face Transformer Reinforcement Learning library for post-training methods such as SFT, DPO, PPO and GRPO on top of transformers.
- turbopuffer
Vector and full-text search engine built from first principles on object storage, aimed at low cost and high scale.
- Tuskira AI Agent Gateway (beta)
Open-source, self-hosted gateway in front of MCP tool servers and LLM APIs: authenticates every call, injects credentials, scopes tools per agent profile and logs it all.
- txtai
All-in-one open-source Python framework for semantic search, RAG, LLM pipelines, workflows and agents, with Web and MCP APIs.
- Unity CLI (beta)
Unity's experimental command-line interface for installing Editors and modules, managing projects and driving a running Editor, with an MCP mode for agents.
- Unreal Engine
Epic Games' game engine (current docs: Unreal Engine 5.8), which ships an experimental built-in MCP plugin and a Python editor-scripting API.
- Unreal Python API (beta)
Epic's Python API for scripting the Unreal Editor (content production and pipeline tasks); an experimental feature also used by community MCP servers.
- Unsloth
Open-source LLM fine-tuning library and desktop app for running and training models locally, with a dual Apache-2.0 / AGPL-3.0 license split.
- vLLM
Apache-2.0 high-throughput LLM inference and serving engine with an OpenAI-compatible API, installable via uv or pip.
- Weaviate
Open-source vector database (written in Go) that stores both objects and vectors.
- WorkOS AuthKit for MCP
WorkOS AuthKit guide and support for securing MCP servers that require authentication.
- WorkOS Vault
Developer-friendly enterprise key management (EKM) to encrypt and optionally store tokens, passwords, certificates, files and other customer data.
- XGrammar
Fast portable structured-generation engine for LLM grammars and constrained decoding (MLC AI).
- Zep
Unified context layer that combines business data, documents and conversations into governed context and memory for agents.