Get the multi-aggregate shadow-analysis breakdown

curl --request GET \
  --url https://api.verlon.ai/v1/ab-tests/:id/shadow-analysis \
  --header 'Authorization: Bearer <token>'

{
  "error": "<string>",
  "message": "<string>",
  "details": "<unknown>"
}

Experiments

Get the multi-aggregate shadow-analysis breakdown

Use when: caller wants the multi-aggregate breakdown of shadow results (segmented by topic, by request type, by latency bucket) for deeper analysis than the headline verdict

Returns the five independent aggregates that power the experiment’s Shadow Analysis section: per-rubric-dimension win/lose/tie counts, position-bias counts by confidence bucket + biased rate, daily Wilson-95%-CI win-rate trend, judge cost vs. budget + projected exhaustion days, and per-variant guardrail trip counts. Implementation lives in services/shadow-analysis.ts so Cortex and Einstein agents can read the same aggregates.

GET

ab-tests

:id

shadow-analysis

Get the multi-aggregate shadow-analysis breakdown

curl --request GET \
  --url https://api.verlon.ai/v1/ab-tests/:id/shadow-analysis \
  --header 'Authorization: Bearer <token>'

{
  "error": "<string>",
  "message": "<string>",
  "details": "<unknown>"
}

Authorizations

Authorization

string

header

required

Layer API key (prefixed sk-). Generated from the dashboard under Settings → API Keys, or via layer key create in the CLI.

Response

Five-bucket analysis payload.

Get the shadow-test summary + verdict Promote a completed shadow experiment (apply or graduate to split)

Documentation Index

Authorizations

Response