"We begin by covering how you can: Build and test prototypes of your tools Create and run comprehensive evaluations of your tools with agents Collaborate with agents like Claude Code to automatically increase the performance of your tools We conclude with key principles for writing high-quality tools we've identified along the way: Choosing the right tools to implement (and not to implement) Namespacing tools to define clear boundaries in functionality"
"In computing, deterministic systems produce the same output every time given identical inputs, while non-deterministic systems-like agents-can generate varied responses even with the same starting conditions. When we traditionally write software, we're establishing a contract between deterministic systems. For instance, a function call like getWeather("NYC") will always fetch the weather in New York City in the exact same manner every time it is called."
MCP tools should be developed with iterative prototypes, comprehensive agent-based evaluations, and collaboration with agents to automatically improve tool performance. Tools function as contracts between deterministic systems and non-deterministic agents, requiring different design principles than traditional APIs. Agents may choose between using a tool, answering from knowledge, or asking clarifying questions, and can hallucinate or misuse tools. High-quality tools employ careful selection of which capabilities to implement, clear namespacing, meaningful contextual returns to agents, token-efficient responses, and precise, well-engineered tool descriptions and specifications.
Read at Anthropic
Unable to calculate read time
Collection
[
|
...
]