Tag: mcp
Everything tagged “mcp” — my own articles first, then the AI Radar entries that carry the same tag.
My posts
AI Radar
- · generativeai.pub
Skills vs. MCP vs. RAG vs. Memory: How to Actually Give AI Agents the Knowledge They Need
Distinguish between skills, MCP, RAG, and memory to optimize AI agent knowledge retrieval. Avoid context bloat and learn when to apply each strategy for accurate task resolution.
- · newsletter.posthog.com
What nobody tells you about writing agent skills
Learn how to design effective AI agent skills using progressive disclosure, prevent context rot, and optimize maintenance workflows for better developer productivity.
- · florian.bruniaux.com
Your agent config is already 90% portable. The gap that's left fits in one TOML header.
Learn how to share AI agent configurations across Claude, Codex, and Cursor using portable standards while handling MCP secrets securely with a custom generator.
- · generativeai.pub
The AI Bottleneck Has Moved. Most People Are Still Solving Yesterday’s Problem.
Learn why AI success now depends on context and system design rather than prompt engineering. Explore shifts in model capabilities and workflow optimization strategies.
- · generativeai.pub
This Repo Cut My Agent’s Token Bill by 88% and the Answer Didn’t Change
Discover how the headroom tool reduces LLM token costs by compressing tool outputs and logs before model ingestion using reversible compression techniques.
- · generativeai.pub
I Read Anthropic's Internal Guide on Building Claude Skills. Here's Everything You Need to Know.
Learn how to build skills for Claude using Anthropic's internal guide. Understand the folder structure, YAML frontmatter, and best practices for effective skill creation.
- · storybook.js.org
Storybook MCP for React
Learn how Storybook MCP enhances AI agents with component metadata, enabling them to reuse components, self-verify with tests, and generate UI faster.
- · ejholmes.github.io
MCP is dead. Long live the CLI
This article argues that the Model Context Protocol (MCP) is unnecessary and that CLIs offer a superior, more practical approach for LLMs to interact with tools.