Back to evaluations
Public evaluation

variant_classification_task

Evaluates variant triage accuracy, ACMG code assignment precision, and JSON output formatting.

Evaluation type
task based
Challenge
Build a Variant Functional Effect Predictor with OpenAI Agents SDK
Difficulty
Intermediate
Rigor
Unspecified

Evaluation overview

How the linked challenge is judged: tasks, benchmarks, and criteria count.

Tasks
1
Benchmarks
0
Criteria
0

Task templates

Inputs and expected outputs.

Task 1

variant_classification_task

Evaluates the agent's ability to classify missense variants into ACMG categories.

Input format

JSON containing genomic variant coordinates and reference/alternate alleles.

Output format

JSON containing ACMG classification, pathogenicity score, and evidence codes.