Discover the Best AI Resources

Apify MCP Server

Lets AI agents discover and run Apify Actors as MCP tools for web scraping and automation; supports OAuth/Streamable HTTP, dynamic tool discovery, output-schema inference, and agentic payments (AGI, x402, Skyfire) so agents like Claude, ChatGPT, or VS Code can call ready-made scrapers and automations.

iii

Provides a shared runtime that composes, extends, and observes services in real time by modeling capabilities as discoverable workers, functions, and triggers. It collapses separate integration surfaces (queues, cron, HTTP, observability) into one live catalog so agents and services can call and trace each other immediately.

Sim

Lets teams build, deploy, and manage AI agents from chat, visual workflows, code, knowledge bases, tables, and more than a thousand integrations.

Fli

Provides programmatic access to Google Flights via a Python library, CLI, and an MCP server — enabling assistants and apps to search flights with filters (time windows, cabin, stops, airlines) by reverse‑engineered API rather than HTML scraping.

PentAGI

Runs penetration tests autonomously: a multi-agent system (researcher, developer, executor) plans attacks, writes and runs exploit code, and chains 20+ tools like nmap, metasploit and sqlmap in isolated Docker containers — for authorized testing only.

OM1

Turns camera, audio, LIDAR and web inputs into robot motion, navigation and speech by routing them through pluggable LLMs and VLMs. Hardware-agnostic Go runtime configured via JSON5, with ROS2/Zenoh middleware for real robots and simulators.

Solace Agent Mesh

Builds event-driven multi-agent AI systems that use a Solace event mesh for agent-to-agent messaging, task delegation, and artifact exchange. Emphasizes asynchronous orchestration, plugin-based extensibility, and integrations with LLMs and external systems.

Train LLM From Scratch

Provides end-to-end PyTorch scripts to download/prepare data, implement a transformer from scratch, train LLMs (13M→billion-scale) and generate text. Emphasizes educational clarity and single‑GPU experiments; useful for researchers or hobbyists, but large-scale training still requires substantial compute and engineering.

Rowboat

Connects to Gmail, Calendar, and meeting notes to build a local, Obsidian-compatible Markdown graph it acts on — drafting emails, briefs, and decks. Memory accumulates instead of resetting each session; runs on local or hosted models, extensible via MCP.

UltraRAG

Wires retrievers, rerankers, and generators as standalone MCP servers orchestrated in YAML, so iterative RAG logic fits in dozens of lines instead of glue code. Adds loops, conditional branches, one-command web UIs, and shared evaluation benchmarks.

fast-agent documentation

MCP-native agent framework built around the Model Context Protocol from the start, with end-to-end tested Sampling and Elicitation. Define agents and multi-step workflows in Python, run terminal-first, and swap Anthropic, Google or local models.

UI-TARS Desktop

Drives your computer from natural language: a vision-language model reads raw screenshots and works the mouse and keyboard like a person, controlling any GUI app without APIs or accessibility hooks. Local or remote operator modes on Windows and macOS.