Code available here:
https://github.com/ai-that-works/ai-that-works/tree/main/2025-09-16-coding-agent-tools-bash-vs-mcp

In this conversation, Dex and Vaibhav delve into the intricacies of coding agents, focusing on the debate between using MCP (Model Control Protocol) and Bash for tool integration. They explore the importance of understanding context windows, token management, and the efficiency of using different tools. The discussion emphasizes the significance of naming conventions, dynamic context engineering, and the engineering efforts required to optimize performance. They also share real-world applications, best practices for using MCPs, and engage with the community through a Q&A session.

Chapters
00:00 Introduction to Coding Agents and Tools
02:22 MCP vs. Bash: Understanding the Debate
05:13 Exploring Context Windows and Token Management
08:13 The Role of Naming in Tool Efficiency
10:47 Practical Examples with Claude and Linear
13:51 Engineering Context for Better Performance
16:06 The Importance of Dynamic Context Engineering
18:59 MCP vs. Bash: Pros and Cons
21:25 Best Practices for Using MCPs
24:08 Leveraging Bash for Efficiency
26:53 Final Thoughts on Context Management
29:51 Q&A and Community Engagement