MCP (Model Context Protocol)
The standard protocol for connecting LLMs to external tools. Not a package you install: an interoperability specification. Concrete MCP servers are the actual tools. Before adding an MCP: weigh the token cost. Each MCP server loads its tools into the agent context, a GitHub MCP injects hundreds of tools, and tool selection burns thousands of tokens per prompt. MCP is justified when: (a) 5+ tools to expose, (b) multiple agents share the same integration (e.g., internal database). For 2-5 deterministic tools, a direct API call from the agent code stays faster and cheaper.
Strengths
- Open Anthropic standard, adopted by Claude, Cursor, Windsurf, Gemini CLI
- Clear client/server architecture: one MCP server = one tool, one namespace
- Multiple transports: stdio (local), HTTP SSE (remote), WebSockets (real-time)
- Explosive community in 2026: +10k public MCP servers listed
Limitations
- Large security attack surface: prompt injection via tool descriptions
- No authentication standard yet, each server handles its own auth
- Hard to debug: MCP errors barely readable without dedicated tooling
- Token cost: every active MCP server consumes context to expose its tools
Best for
- AI agents requiring access to APIs, files, databases
- Integrating existing tools (GitHub, Jira, Slack) into an LLM workflow