Press n or j to go to the next uncovered block, b, p or k for the previous block.
| 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 | 1x 1x 1x 1x 1x 1x 1x 1x 1x 1x 1x 1x 1x 1x 1x 1x 1x 1x 1x 1x 1x 1x 1x 1x 1x 1x 1x 1x 1x 1x 1x 1x 1x 1x 1x 1x 1x 1x 1x 1x 1x 1x 1x 1x 1x 1x | import { EvaluationMethodName } from '@shared/types/evaluations';
import { z } from 'zod';
/**
* Display information for UI presentation
* Contains label-value pairs or plain text sections
*/
export const EvaluationDisplayInfo = z.object({
/** Main label for this display item (e.g., "Verdict", "Reasoning", "Performance") */
label: z.string(),
/** Plain text content to display */
content: z.string(),
});
export type EvaluationDisplayInfo = z.infer<typeof EvaluationDisplayInfo>;
export const SkillOptimizationEvaluationResult = z.object({
evaluation_id: z.uuid(),
method: z.enum(EvaluationMethodName),
score: z.number().min(0).max(1),
extra_data: z.record(z.string(), z.unknown()),
/** Standardized display information for UI presentation */
display_info: z.array(EvaluationDisplayInfo),
/** The name of the model used for judging (null for non-LLM evaluations) */
judge_model_name: z.string().nullable().optional(),
/** The provider of the model used for judging (null for non-LLM evaluations) */
judge_model_provider: z.string().nullable().optional(),
});
export type SkillOptimizationEvaluationResult = z.infer<
typeof SkillOptimizationEvaluationResult
>;
/** An optimized configuration used for AI inference.
*
* A group of logs were generated using a clustering algorithm
* to group similar logs together.
* Then this **optimal** configuration was generated to produce
* the best results on one of the clusters.
* We assume this configuration is also optimal
* for any future requests that are similar to any of the
* logs in the cluster. */
export const SkillOptimizationEvaluationRun = z.object({
id: z.uuid(),
agent_id: z.uuid(),
skill_id: z.uuid(),
cluster_id: z.uuid().nullable(),
log_id: z.uuid(),
/** The results of when the arm pull was evaluated */
results: z.array(SkillOptimizationEvaluationResult),
/** When the evaluation run was created */
created_at: z.iso.datetime({ offset: true }),
});
export type SkillOptimizationEvaluationRun = z.infer<
typeof SkillOptimizationEvaluationRun
>;
export const SkillOptimizationEvaluationRunQueryParams = z
.object({
id: z.uuid().optional(),
agent_id: z.uuid().optional(),
skill_id: z.uuid().optional(),
cluster_id: z.uuid().optional(),
log_id: z.uuid().optional(),
created_after: z.string().datetime().optional(),
created_before: z.string().datetime().optional(),
limit: z.number().min(1).max(100).optional(),
offset: z.number().min(0).optional(),
})
.strict();
export type SkillOptimizationEvaluationRunQueryParams = z.infer<
typeof SkillOptimizationEvaluationRunQueryParams
>;
export const SkillOptimizationEvaluationRunCreateParams = z
.object({
agent_id: z.uuid(),
skill_id: z.uuid(),
cluster_id: z.uuid(),
log_id: z.uuid(),
results: z.array(SkillOptimizationEvaluationResult),
})
.strict();
export type SkillOptimizationEvaluationRunCreateParams = z.infer<
typeof SkillOptimizationEvaluationRunCreateParams
>;
|