AIHero

    Tools & Environment

    Tool call

    The model's output naming a tool and its arguments — just structured text. The harness has to read it and execute.

    Matt Pocock
    Matt Pocock

    The model's output naming a tool and its arguments — just structured text. It doesn't do anything on its own; the harness has to read it and execute. Produced by the model in one model provider request.

    The lifecycle of a tool call:

    StepWhoWhat happens
    1ModelLearns which tools exist from descriptions in the system prompt
    2ModelEmits a call — tool name plus arguments, usually JSON — and stops
    3HarnessParses the call and checks it against the permission mode
    4HarnessExecutes it if allowed
    5HarnessSends the outcome back as a tool result in the next request

    One turn of agent work is usually many of these round trips chained together.

    Because the call is generated by next-token prediction like everything else, it can be wrong the way any model output can be wrong: a path that doesn't exist, a flag the command doesn't have, arguments that are plausible rather than correct. The harness executes what was written, not what was meant — a mistyped path doesn't error gracefully, it edits the wrong file.

    Usage:

    "It said it ran the tests but the file timestamps haven't changed."

    "Look at the transcript — did it actually emit a tool call, or just describe running them? The model produces the call, but if the harness didn't execute it, nothing happened."

    Want more than vocabulary?

    Join AI Hero for practical skills, thinking on AI engineering, and resources that keep you ahead of the curve.

    I respect your privacy. Unsubscribe at any time.

    Share