SelfExplainInclude
Interface: SelfExplainInclude
Defined in: src/lib/trace-toolpack/selfExplain.ts:72
How much of a turn's evidence the binding keeps.
Both default to TRUE, which is the point: the tools that read them
(read_narrative, inspect_tool_call) are on the catalog either way,
and a tool that answers "⚠ no evidence" by default is a tool that
teaches the model not to call it. Turn one off when the cost matters
more than the answer — a very long-running turn, or a run whose
narrative would repeat what the structured tools already say.
Properties
events?
readonlyoptionalevents?:boolean
Defined in: src/lib/trace-toolpack/selfExplain.ts:80
A bounded tail of the run's typed events → tool-call timings and
outcomes in inspect_tool_call. Default true. Off means no wildcard
event subscription is made at all, not a subscription that is ignored.
narrative?
readonlyoptionalnarrative?:boolean
Defined in: src/lib/trace-toolpack/selfExplain.ts:74
The run's plain-English story → the read_narrative tool. Default true.