Back to evaluations
Public evaluation

AutomatedContractNegotiation

Evaluation will focus on the negotiation system's ability to successfully reach agreements, adhere to RSL 1.0 rules, and record transactions securely via MCP. The quality of agent decision-making and negotiation strategies will be key.

Evaluation type
task based
Challenge
AI Content Licensing Agent
Difficulty
Advanced
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

AutomatedContractNegotiation

Simulate multiple negotiation rounds between a Publisher Agent and a Crawler Agent based on a given content asset and RSL 1.0 rules. The goal is to reach a signed agreement or log a failed negotiation.

Input format

{'content_asset_id': 'abc123', 'publisher_rules': {'min_price_per_token': 0.001, 'licenses_available': ['commercial', 'research']}, 'crawler_interest': {'tokens_desired': 100000, 'license_type': 'commercial', 'max_bid_per_token': 0.0015}}

Output format

{'negotiation_id': '...', 'status': 'success'|'failure', 'final_agreement': {'content_asset_id': '...', 'license_type': 'commercial', 'price_per_token': 0.0012, 'total_cost': 120.0, 'signed_by_publisher': true, 'signed_by_crawler': true}, 'negotiation_log': [...], 'mcp_transaction_id': '...'}