Documentation

Evaluation Dashboards

EvaluationDashboardsResource

Methods

create() ->
EvaluationDashboard
post/v5/evaluation-dashboards

Create a new evaluation dashboard for an evaluation or evaluation group

list() -> SyncCursorPage[
EvaluationDashboard
]
get/v5/evaluation-dashboards

List dashboards filtered by evaluation_id, evaluation_group_id, tags, creators, or search

retrieve(, ) ->
EvaluationDashboard
get/v5/evaluation-dashboards/{dashboard_id}

Get a single evaluation dashboard by ID

update(, ) ->
EvaluationDashboard
patch/v5/evaluation-dashboards/{dashboard_id}

Partially update dashboard metadata (name, description, widget_order)

archive() ->
EvaluationDashboard
delete/v5/evaluation-dashboards/{dashboard_id}

Soft delete an evaluation dashboard

Domain types

class EvaluationDashboard: ...
Evaluation Dashboards

Widgets

EvaluationDashboardsResource.WidgetsResource

Methods

create(, ) ->
EvaluationDashboardWidgetWithResult
post/v5/evaluation-dashboards/{dashboard_id}/widgets

Create a new widget, add it to the dashboard, and compute its results

update(, ) ->
EvaluationDashboardWidgetWithResult
patch/v5/evaluation-dashboards/{dashboard_id}/widgets/{widget_id}

Update a widget and compute its results. If the widget is only used by this dashboard, it is updated in place. If shared across multiple dashboards, a copy is created.

remove(, )
delete/v5/evaluation-dashboards/{dashboard_id}/widgets/{widget_id}

Remove a widget from the dashboard (does not delete the widget)

Domain types

class EvaluationDashboardWidget: ...
class EvaluationDashboardWidgetResult: ...
class EvaluationDashboardWidgetResultResponse: ...

Computed result for a widget - used in widget creation response

class EvaluationDashboardWidgetWithResult: ...

Response model for widget creation - includes widget and computed result

Literal["bar", "histogram", "donut", "scatter", "metric", "table", "markdown", "heading", "timeseries"]

Widget types for dashboard visualizations

class Filter: ...

Filter clause with conditions connected by logical operators.

Conditions are evaluated left-to-right without precedence (no nesting/parentheses). Example: condition1 AND condition2 OR condition3 evaluates as ((condition1 AND condition2) OR condition3)

Example: { "conditions": [ {"column": "score", "operator": ">", "value": 0.5}, {"column": "category", "operator": "=", "value": "test"} ], "logicalOperators": ["AND"] }

class MetricQuery: ...

Query that returns a single metric value (used for metric widgets).

Used for widget type: metric. Enforces exactly 1 aggregation in select. Returns: {"type": "metric", "data": ...}

Example SQL equivalent: SELECT AVG(score) as average_score FROM evaluation_items

class SelectItem: ...

Column in SELECT clause

class SeriesQuery: ...

Query that returns a series of records (used for table/bar/histogram/donut/scatter widgets).

Used for widget types: table, bar, histogram, donut, scatter. Returns: {"type": "series", "data": [...]}

Example SQL equivalent: SELECT category, AVG(score) as avg_score, COUNT(*) as count FROM evaluation_items WHERE score > 0.5 AND category = 'test' GROUP BY category ORDER BY avg_score DESC LIMIT 100