ML//agent//agent harness//tool interface

When a program calls an API, its implementation and interface can be inspected separately. Agent tools split even more sharply: the model normally sees only a name, description, input schema, examples, and previous outputs.

When a program calls an API, its implementation and interface can be inspected separately. Agent tools split even more sharply: the model normally sees only a name, description, input schema, examples, and previous outputs.

The hidden implementation might use ripgrep, a vector index, Tree-sitter, or another model. If each returns compatible search results, the LLM may operate through the same abstraction without knowing what changed underneath.

Names and parameter labels therefore create affordances. execute_operation is technically flexible but behaviorally vague; search_payment_logs makes its intended use legible. A capable tool with a bad interface can be functionally absent because the model rarely selects it correctly.

Tools are APIs for software and instructions for a probabilistic policy at the same time.