When does MCP make sense vs CLI?

kept by eddie mar 02

The Hacker News discussion weighs the trade-offs between MCP and CLI for AI agent tool calling, finding that each has valid use cases depending on context.

A Year Of Vibes
Armin Ronacher reflects on 2025 as a year of agentic coding tools like Claude Code, which replaced much of his direct programming, and discusses the cultural, technical, and social challenges this shift presents for software engineering.
also on Amp, pi, Claude Code, #software-engineering, #developer-tools
The Coding Agent Is Dead - Amp
Amp is killing its editor extensions for VS Code and Cursor to move to a CLI-only product, arguing new AI models no longer need the hand-holding.
also on Amp, #cli, #ai-agents, #developer-tools
Clopus-Watcher: An autonomous monitoring agent
The author demonstrates Claude Code running in a Kubernetes cronjob as an autonomous on-call agent that monitors pods, applies hotfixes, and documents incidents.
also on Kubernetes, Claude Code, #ai-agents, #developer-tools
GitHub - clawkwork/clawk at console.dev
clawk gives coding agents their own disposable Linux VM, letting them install packages and run code while keeping host files, keys, and network strictly isolated.
also on Kubernetes, Claude Code, #software-engineering, #developer-tools
MCP is dead. Long live the CLI
The Model Context Protocol (MCP) offers no real benefit over command-line interfaces (CLIs), which are more composable, debuggable, and reliable for LLM tool use.
also on pi, OpenClaw, #cli, Claude Code, #software-engineering, #developer-tools