GitHub hosts many curated "awesome" lists that aggregate tools, libraries and resources across topics relevant to developers and AI practitioners.

Overview

These collections save time and help teams discover free services, APIs and agent frameworks without exhaustive searching or filtering online.

Key lists

Below are twelve repositories selected for developers, AI builders and teams seeking hosted services, agent tooling and self-hosted alternatives today.

• awesome (481k+) compiles index pages that point to topic-specific awesome lists, covering languages, tools, design and security resources and more.
• free-for-dev lists permanently free tiers for hosting, databases, email, monitoring and CI/CD, excluding temporary trials for teams.
• public-apis catalogs 1.5K publicly accessible APIs by category, including weather, finance, images, games and news endpoints for projects.
• awesome-selfhosted collects open-source alternatives to commercial services, offering replacements for products like Notion, Google Photos, Zapier and many more.
• skills-anthropic is the official Anthropic repository of skills and integrations recommended for use with their cloud subscription provider and related tooling.
• awesome-claude-code gathers skills, hooks, slash commands, orchestrators and plugins developed for Claude Code integrations, with practical examples for integration today.
• awesome-mcp-servers indexes thousands of MCP servers that enable agents to connect to browsers, databases and spreadsheet services remotely and securely.
• awesome-codex-skills curates skills for analytics, communication, writing, productivity and development tailored to Codex-based agents, with examples and guides.
• awesome-codex-subagents lists more than 130 subagents for specific tasks, with examples and out-of-the-box usage patterns available for adaptation.
• awesome-agent-orchestrators highlights tools for running multiple coding agents in parallel such as claude-squad, crystal and vibe-kanban among dozens of other orchestrators.
• awesome-llm-apps contains 100+ ready AI agents and RAG applications with full source code demonstrating retrieval-augmented generation patterns and deployment examples.
• system-prompts-and-models-of-ai-tools collects leaked and public system prompts for Cursor, Devin, Lovable, v0, Claude Code and other models as reference material.

How to use them

If you are new, avoid installing everything at once; prioritize repositories that solve immediate problems for your projects and teams.

A practical approach is to have your agent scan selected lists and extract relevant tools, integrations and setup steps tailored to your specific workflows and constraints.

These curated indexes reduce discovery time and can help you prototype faster while avoiding unnecessary costs on services you do not require.