🧠 MCP vs Agent Skills: Capabilities and Procedures Explained
MCP and Agent Skills are complementary primitives. MCP provides capabilities (what agents can do). Skills provide procedures (how agents should approach it).
Agents and the protocols that connect them.
MCP and Agent Skills are complementary primitives. MCP provides capabilities (what agents can do). Skills provide procedures (how agents should approach it).
Everyone talks about MCP Tools, but Resources solve a different problem entirely. Learn why application-controlled context matters and why host support is holding back this powerful primitive.
Your MCP tool descriptions are eating your context window. Learn about the token efficiency problem and solutions emerging at server, protocol, and host layers.
Applying Roy Fielding's architectural methodology to analyze Model Context Protocol (MCP), examining its constraints, trade-offs, and security implications.
OpenAI has announced full support for MCP across its product line, marking a significant consolidation around standards in the AI industry.
APIs are the connective tissue of our digital world. Now there's a new protocol reshaping the landscape: Model Context Protocol (MCP). Let's debunk the misconception that MCP will kill APIs.
APIs provide the cement that binds us to our digital lives. For those building these connective tissues, choosing the right tools to describe and document APIs isn't just a technical decision—it's strategic.
Just as the 1984 Macintosh GUI revolutionized computing, AI agents are creating an even more exciting transformation, making technology more accessible, intuitive, and powerful than ever before.
Need help with your API strategy?
15+ years of experience in API governance, product strategy, and go-to-market.