Filtered to AI Infrastructure · show all articles
Tool protocols standardize how a model asks a server to act. They say far less about what happens when a schema changes underfoot, a timeout fires before a write commits, a token expires mid-task, or a session drops — production finds every one of those gaps.
Four calling conventions, each dated to a real announcement: a manifest borrowed from web APIs, a schema built into the model call itself, incompatible framework wrappers, and a standard now governed by the vendors who once competed on incompatibility.
Inside a language model's context window, a tool description, a tool result and a resource all read the same as an instruction — which means a compromised or malicious MCP server can write the agent's next action itself.
A model that only emits text cannot query a database or send an email. Four documented patterns cross that boundary, and each one answers a different question — none is a smaller version of another.
A tool schema is a contract offered to a caller that cannot ask a follow-up question mid-call. This is a practitioner's guide to writing that contract so it survives retries, versioning, and a model that will eventually call it wrong.