# Resource Summary Link
1 MCP official docs – Prompts section Spec page explaining how MCP prompts work, with runnable JSON examples  https://modelcontextprotocol.io/docs/concepts/prompts
2 YouTube: “Building AI Agents With MCP: A Complete Walk-through” 30-minute screen-share that spins up a server and walks through an end-to-end agent flow  https://www.youtube.com/watch?v=9izHUWherYw
3 Bitcot blog: “How to Build AI Agents Using MCP – A Hands-On Guide” Step-by-step article that scaffolds an MCP agent, with copy-pasteable code blocks  https://www.bitcot.com/how-to-build-ai-agents-using-mcp-a-complete-guide/
4 Composio blog: “The Complete Guide to Building MCP Agents” Detailed tutorial using OpenAI SDK + Composio to construct and deploy agents  https://composio.dev/blog/the-complete-guide-to-building-mcp-agents
5 DEV series (Part 10): “Sampling & Prompts in MCP” Deep-dive post on prompt workflows and safety hooks inside MCP agents  https://dev.to/alexmercedcoder/a-journey-from-ai-to-llms-and-mcp-10-sampling-and-prompts-in-mcp-making-agent-workflows-2446
6 GitHub repo: lastmile-ai/mcp-agent Minimal open-source framework showing how to wire an MCP server and custom agent  https://github.com/lastmile-ai/mcp-agent
7 Medium: “AI Agents Using MCP = Your ‘Guy in the Chair’” Concise conceptual primer with design heuristics and sample code snippets  https://buckhouse.medium.com/ai-agents-using-mcp-your-guy-in-the-chair-f32ac850c60e
8 Reddit thread: “MCP clients – best practices for prompts” Community cheat-sheet of prompt patterns and real-world gotchas for clients  https://www.reddit.com/r/mcp/comments/1lnf32b/mcp_clients_best_practices_for_prompts/