pj ec5872ac6e feat(trace): record one typed outcome per model-driven step ...

llm-calls.jsonl carries the prompts as sent, the candidate list as the model
saw it, the screenshot reference, the raw response, tokens, latency and how the
step ended. It sits beside trace.jsonl rather than inside it because every trace
line already carries a full hierarchy and both the replay server and the
campaign summarizer scan all of them; folding prompts in would grow the lines
those readers parse for data neither reads.

Claude-Session: https://claude.ai/code/session_01A5KmftdEJ49A9z5mF5ESrX

2026-08-12 22:56:58 +05:30

..

llm_call_test.go

feat(trace): record one typed outcome per model-driven step

2026-08-12 22:56:58 +05:30

llm_call.go

feat(trace): record one typed outcome per model-driven step

2026-08-12 22:56:58 +05:30

writer_test.go

experiment instrumentation: step budgets, arm labels, campaign runner (#72)

2026-08-12 22:20:31 +05:30

writer.go

feat(trace): record one typed outcome per model-driven step

2026-08-12 22:56:58 +05:30