Skip to content
The Test Cabinet The Test Cabinet

One-shot

One-shot (slug one-shot) is the default orchestrator: a single harness session driven to completion. The harness is handed the test case’s prompt and its own agent loop runs to completion — that is the whole strategy. It reproduces the original single-session behaviour exactly, so a one-shot run’s metrics are identical to a run with no orchestration layer at all.

Runner

Section titled “Runner”

The runner reads no parameters. It is a single invocation of the tcab-session wrapper against the goal carried in TCAB_PROMPT:

Terminal window
exec tcab-session "$TCAB_PROMPT"

tcab-session runs the selected harness’s CLI with that harness’s exact session arguments, substituting the prompt — so the runner needs to know nothing harness-specific.

Parameters

Section titled “Parameters”

None. A multi-session strategy such as Ralph Loop declares a [params] table; one-shot does not.