Agent Evals

Create Benchmark

Create an empty benchmark.

post/api/v0/agent/evals/benchmarks

Request

Base URL
https://api.outerport.com (the document declares no server; this is the origin it was published from)
Auth
HTTP bearer
TypeScript
https://skmtc.dev/outerport/apis/outerport-api/docs/api/v0/agent/evals/benchmarks/post.ts
A function with typed inputs. Its response is checked against the schema.

namestring required

The benchmark's name.

persona_idstring uuid nullable

The persona runs use unless they name another. Null for the default persona.

rubricstring

Instructions the judge grades every case by.

Successful Response

idstring uuid required

The ID of the benchmark.

namestring required

The benchmark's name.

persona_idstring uuid nullable

The persona a run uses unless it names another. Null for the default.

rubricstring required

Instructions the judge grades every case by.

created_atstring date-time required

When the benchmark was created.

updated_atstring date-time required

When the benchmark or its cases last changed.