To avoid overwhelming an LLM's context with hundreds of tools, a dynamic MCP approach offers just three: one to list available API endpoints, one to get details on a specific endpoint, and one to execute it. This scales well but increases latency and complexity due to the multiple turns required for a single action.

AI plugins (MCPs) constantly consume valuable context window space, even when not in use. Integrating tools via Command-Line Interfaces (CLIs) is more efficient. The AI can execute local CLI commands as needed, providing full tool functionality without the persistent context overhead.

The vision for Model Context Protocol (MCP) is to let AIs perform complex, multi-app tasks. However, translating a full API like Stripe's into MCP tools overwhelms current models' context windows, making them confused and ineffective. This forces developers to handcraft a small subset of tools.

To overcome LLM limitations, successful Model Context Protocol (MCP) design involves severe constraints: keep the number of tools low, use precise yet concise names and descriptions, minimize input parameters, and return only essential data. This handcrafted approach is necessary for models to perform reliably.

Instead of direct API calls, build Model-Controlled Program (MCP) servers. They act as better guardrails for the AI, allowing it to interact with external data more effectively and even suggest novel use cases based on API documentation.

Users often fail with MCP by expecting it to handle complex workflows instead of simple tool interactions. A key mistake is connecting too many irrelevant servers, which pollutes the AI's context window with unused tool descriptions and degrades performance. Keep the toolset minimal and relevant to the task.

Instead of giving an LLM hundreds of specific tools, a more scalable "cyborg" approach is to provide one tool: a sandboxed code execution environment. The LLM writes code against a company's SDK, which is more context-efficient, faster, and more flexible than multiple API round-trips.

"Code Mode" is not an alternative to MCP but a more efficient way to use it. Instead of multiple sequential tool calls, the model generates a single script that executes multiple actions in a sandbox. MCP still provides the core benefits of authentication, discoverability, and a standardized, LLM-friendly API.

To solve the problem of MCPs consuming excessive context, advanced AI clients like Cursor are implementing "dynamic tool calling." This uses a RAG-like approach to search for and load only the most relevant tools for a given user query, rather than pre-loading the entire available toolset.