Trace event schema

Each session is one JSON Lines file. Each line is one event. Every event carries v (schema version), session_id, ts (ISO 8601, UTC), and type.

session_start
  cwd, base_commit, remote_url_hash, model, harness, harness_version, source,
  task_id, config_hash

message
  turn, role (user | assistant), text_redacted, text_hash, truncated (bool)

tool_call
  turn, tool_name, tool_use_id, input_redacted, input_hash,
  output_redacted, output_hash, truncated (bool), duration_ms

oracle_verdict
  attempt, mode (baseline | grade), verdict (PASS | FAIL | ERROR),
  tests_hash, image_digest, diff_hash, oracle_host, signature (optional)

session_end
  outcome (flipped | no_flip | no_task | aborted), attempts,
  final_diff_hash, reason (from the harness), ended_at

Rules: text_redacted, input_redacted, and output_redacted are the only fields that ever hold content, and they hold it after redaction. The _hash fields are SHA-256 of the unredacted value, so a later pass can tell whether two truncated values were the same without storing them. remote_url_hash rather than the URL, because remote URLs sometimes carry credentials. The verdict's signature is over the canonical JSON of the other verdict fields, with the oracle host's key.

@incollection{anderson2026continuousdeliveryof,
  author    = {Anderson, Mac},
  title     = {Trace event schema},
  booktitle = {Building a continuous delivery system for fine-tuned open-source, open-weight models trained on your organization's traces},
  year      = {2026},
  url       = {https://macanderson.com/research/continuous-delivery-of-fine-tuned-open-weight-models/trace-event-schema}
}