Skip to main content
GET
Get evaluation overview

Authorizations

Authorization
string
header
required

A Recursion API key, created in the console under API keys.

Query Parameters

range
enum<string>
default:28d

Length of the current window. The prior window is the adjacent interval of equal length.

Available options:
7d,
28d,
90d
evaluator_agent_version_id
string<uuid>
required

Immutable evaluation-agent version whose rubric and verdicts define this overview.

target_agent_id
string<uuid>

Optionally narrow page-level metrics, table rows, and retained history to one target agent.

series_target_agent_id
string<uuid>[]

Repeat to explicitly compare up to five target agents; absent values select the page target or the first five whole-filter agents.

Maximum array length: 5

One canonical UUID identifying an organization-scoped comparison agent or immutable version.

series_previous_target_agent_version_id
string<uuid>[]

Repeat to add scoped previous target-agent versions; automatic latest versions plus these values may total at most twelve traces.

Maximum array length: 11

One canonical UUID identifying an organization-scoped comparison agent or immutable version.

series_metric
enum<string>
default:overall

Verdict population plotted for each selected target-agent version.

Available options:
overall,
criterion
series_criterion_key
string

Rubric key plotted when series_metric is criterion; required exactly for that metric.

series_resolution
enum<string>

Opt into bounded adaptive chart positions from exact timestamps through days. Omit for the legacy fixed daily grid.

Available options:
adaptive
limit
integer
default:25

Maximum target-agent rows to return. Defaults to 25.

Required range: 1 <= x <= 100
page_token
string

Opaque signed continuation bound to the organization, frozen watermark, range, evaluator, singular table target, and effective limit. Comparison agents, versions, and metric may change without resetting the table page. Expires after one hour.

Response

Complete fixed-window evaluation Overview plus an optional signed target-agent continuation.

Complete fixed-window evaluation Overview plus an optional signed target-agent continuation.

agent_rows
object[]
required

Ranked target-agent metric rows for this signed cursor page.

as_of
string<date-time>
required

Durable analytics watermark freezing every query in this response.

change_markers
object[]
required

Immutable agent-version and model markers when one target agent is selected.

change_markers_status
enum<string>
required

Whether markers are complete or require selecting one target agent.

Available options:
complete,
select_target_agent
comparison
object
required

Bounded multi-agent and immutable-version traces for one selected metric.

Example:
criterion_columns
string[]
required

Stable key order used to align every agent row's criterion_metrics array.

criterion_series
object[]
required

Aligned criterion verdict series for the complete frozen window; legacy responses use a daily grid.

from
string<date-time>
required

Inclusive UTC start of the current Overview window.

prior_from
string<date-time>
required

Inclusive UTC start of the adjacent equal-length prior window.

prior_to
string<date-time>
required

Exclusive UTC end of the prior window, equal to from.

tiles
object
required

Fixed headline coverage, verdict, criterion, and exact-cost metrics.

Example:
to
string<date-time>
required

Exclusive UTC end of the current window, equal to as_of.

criterion_series_granularity
enum<string>

Adaptive position semantics: exact timestamps are criterion-local, while interval granularities use shared aligned positions; omitted for the legacy daily response and empty adaptive collections.

Available options:
exact,
second,
minute,
hour,
day
next_page_token
string

Signed one-hour continuation for the next ranked target-agent page at the same watermark.