Skip to main content
Back to challenges
Challenge

Vercel AI SDK & Agno Subcontractor General Liability Risk Classification System

Court decisions on general contractor liability for subcontractor injuries impact commercial liability underwriting. Build a TypeScript application using Vercel AI SDK paired with Agno agent microservices to parse subcontractor labor agreements and classify indemnification liability risk.

Machine LearningHosted by Vera
Sign in free to start
Challenge brief

What you are building

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

Develop a TypeScript and Python service using Vercel AI SDK and Agno to audit subcontractor liability contracts.

Delivery guide

How work is evaluated

Evaluation

Evaluates contract legal risk identification accuracy and latency of streaming output.

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

enum_validity

Ensures liability_rating is one of LOW_RISK, MEDIUM_RISK, HIGH_RISK

Binary check

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

Dimension 2

clause_detection_accuracy

Accuracy in detecting missing legal clauses • target: 0.9 • 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

  • Build AI SDK streaming responses for interactive contract risk analysis

  • Construct Agno agent endpoints for legal clause classification

  • Detect missing cross-indemnification and additional insured endorsements

  • Provide risk scoring and compliance feedback to commercial brokers

Resources and assets

Reference links and supporting material

Dataset notes

Standard US general contractor and subcontractor agreements with annotated liability provisions.

Challenge setup

Evaluates contract legal risk identification accuracy and latency of streaming output.

Challenge input

JSON containing subcontractor contract text snippet.

AI SDK

Unified streaming and model execution interface

Agno

Specialized agent backends for legal classification

Evaluated output

JSON containing liability_rating, missing_clauses list, and GC_protection_valid boolean.

Evaluation checks
  • Ensures liability_rating is one of LOW_RISK, MEDIUM_RISK, HIGH_RISK
  • Accuracy in detecting missing legal clauses • target: 0.9 • range: 0-1
Configured evaluation evidence
  • Benchmark: ContractRiskBench
  • Clause Detection Accuracy target: 0.9
  • 1 public reference case
Configured runtime
  • JavaScript execution harness
  • javascript 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
  • Agno · Required
  • Vercel AI SDK · Optional
  • OpenAI · Optional
Orchestration
  • Agno · Required
  • Vercel AI SDK · Optional

Evaluation contract

  • enum_validity · Weight 1
  • clause_detection_accuracy · Weight 1

Metrics

  • clause_detection_accuracy · Target 0.9

Benchmarks

  • ContractRiskBench
  • general

Reference cases

  • 1 public reference case

Workflow tasks

  1. Task 1

    Input: JSON containing subcontractor contract text snippet.

    Output: JSON containing liability_rating, missing_clauses list, and GC_protection_valid boolean.

    Execution harness: JavaScript

Evaluation tools

  • AI SDK · TypeScript AI Framework
  • Agno · Agent Engine

Recipe freeze marker

No recipe freeze time is recorded.

Frequently Asked Questions about Vercel AI SDK & Agno Subcontractor General Liability Risk Classification System