Agents
MCP tool design: Practical approaches and tradeoffs
MCP tool design involves deciding the granularity of tools—whether they map to individual API calls or complete workflows—which directly impacts how many tools agents need and their effectiveness. Eff
MCP tool design involves deciding the granularity of tools—whether they map to individual API calls or complete workflows—which directly impacts how many tools agents need and their effectiveness. Effective design requires balancing tool quantity: too few tools limit the LLM's ability to gather necessary context, while too many can cause confusion and hallucinations. Tool descriptions are critical decision surfaces for LLMs—a 2025 study found 97.1% of MCP tool descriptions have quality issues, and clear descriptions are essential since they guide the LLM's tool selection process.
Source: AWS ML Blog | 2026-07-09