ReadonlycwdReadonlyexitReadonlyhooksThe hooks that fired during the run, each with its decision — parsed from
the CLI's hook_response stream events. Same capture requirement as
toolCalls (empty without the stream). Lets a test assert hook firing
honestly instead of via a marker file.
ReadonlymodelThe requests the model received, captured by the scripted mock — each with
its system prompt and messages, flattened to text. Lets a test assert
what actually reached the model (a SessionStart hook's injected context, a
slash command's expansion), not just that a hook fired. Harness tier
only: the mock sees the requests, so this is populated by runHarnessTest
(with or without transcript); the eval tier drives the real API, so its
modelRequests is always empty.
ReadonlyoutputThe agent's final answer text (the terminal result event), or "".
Optional ReadonlyrepliesEvery reply the agent ended a turn with, in order — the output is only
the last. A run a Stop hook continued has two, and anything the first one
printed (a status footer) shows only here. Text the agent writes before a
tool call is narration, not a reply. Absent when the harness's run output
does not tell replies apart, or the run was not streamed (transcript: true).
ReadonlystdoutOptional ReadonlystubEvery invocation of a tool stub during the run, in order — including calls a
script the agent ran made itself (execFile(gh, …)), which never appear as
a tool call. Each says which rule answered it, or that none did. Absent when
the run had no stubs (eval tier: stubs on the spec).
Optional ReadonlysubagentsSub-agent (Task) runs as nested traces, keyed by subagent_type. A
subagent runs its own session; CC tags its events with parent_tool_use_id
(= the Task tool call) so its tool calls are recovered into a sub-trace
here, lettng a test assert what the subagent DID (not just that Task fired).
Empty unless the stream was captured / the harness emits subagent events.
ReadonlytoolThe tools the agent invoked, each paired with its result (parsed from the stream).
Readonlyturnsnum_turns reported by claude, or 0.
ReadonlyusageCost / latency / tokens for this run (use as metrics, e.g. { cost: ctx.usage.costUsd }).
Context handed to
measureafter a run, to compute that run's metrics. It is aTrace(so the bare predicatesusedTool/skillResolved/toolCount/toolUsedWithfromharness-assert.tsrun over it, the same as over arunHarnessTestresult) plus the eval-onlyshend-state probe andusage.