Skip to main content
Back to challenges
Challenge

Build a Claude Agents SDK Microdrama Content Moderation System

Short-form video and microdrama platforms experience high content upload volumes requiring rapid policy moderation. Build a Claude Agents SDK video transcript moderation agent that analyzes episode scripts and metadata to flag regulatory policy breaches with 95% classification precision.

Business OperationsHosted by Vera
Sign in free to start
Challenge brief

What you are building

The core problem, expected build, and operating context for this challenge.

Construct a Claude Agents SDK content moderation system analyzing video transcripts and episode tags for streaming compliance.

Delivery guide

How work is evaluated

Evaluation

Evaluates policy flag precision and age rating assignment accuracy.

Datasets

Challenge datasets

Public datasets, plus the private datasets you have access to.

Evaluation rubric

How submissions are scored

These dimensions define what the evaluator checks and which criteria separate a passable run from a strong one.

Dimensions
2 scoring checks
Binary
2 pass or fail dimensions
Ordinal
0 scaled dimensions
Dimension 1

policy_flag_check

Verifies content violation flag generation

Binary check

This dimension contributes its full weight only when the submission satisfies the requirement. Partial credit is not awarded.

Dimension 2

classification_precision

Precision of assigned policy categories • target: 0.95 • range: 0-1

Binary check

This dimension contributes its full weight only when the submission satisfies the requirement. Partial credit is not awarded.

Learning goals

What you should walk away with

  • Utilize Claude Agents SDK for structured multi-category content classification

  • Extract timestamped policy violation markers from media transcripts

  • Apply age rating rules based on regional broadcasting standards

  • Generate automated compliance reports for digital video publishers

Resources and assets

Reference links and supporting material

Dataset notes

Synthetic microdrama transcripts with annotated content violations and timestamp markers.

Challenge setup

Evaluates policy flag precision and age rating assignment accuracy.

Challenge input

JSON transcript script object with timestamp markers

Claude Agents SDK

Agent framework for deep script analysis

Evaluated output

JSON compliance summary with rating and policy_flags

Evaluation checks
  • Verifies content violation flag generation
  • Precision of assigned policy categories • target: 0.95 • range: 0-1
Configured evaluation evidence
  • Benchmark: Media Content Policy Benchmark
  • Classification Precision target: 0.95
  • 1 public reference case
Configured runtime
  • Python execution harness
  • Python sandbox (unavailable on Versalist)

Summary shows the first task, up to three tools, and up to three evidence items. Open the technical recipe to view all configured items.

View technical recipe

Configured tools

Action Space
  • Confident AI · Required
  • Hamming · Optional
  • Meta · Optional

Evaluation contract

  • policy_flag_check · Weight 1
  • classification_precision · Weight 1

Metrics

  • classification_precision · Target 0.95

Benchmarks

  • Media Content Policy Benchmark
  • general

Reference cases

  • 1 public reference case

Workflow tasks

  1. Task 1

    Input: JSON transcript script object with timestamp markers

    Output: JSON compliance summary with rating and policy_flags

    Execution harness: Python

Evaluation tools

  • Claude Agents SDK · Agent Framework

Recipe freeze marker

No recipe freeze time is recorded.

Frequently Asked Questions about Build a Claude Agents SDK Microdrama Content Moderation System