Challenge

AI Content Licensing Agent

To establish rules for AI web crawlers, this challenge tasks you with building an advanced A2A multi-agent system using LangGraph and Marvin to automate content licensing negotiations between 'Publisher Agents' and 'Crawler Agents.' Each agent type will be powered by Claude Opus 4.5, leveraging its sophisticated reasoning for contract understanding and negotiation. The system must implement graph-based workflows to manage negotiation states, enforce content rules (simulated), and ensure secure, auditable transactions via an MCP for contract finalization and compensation. This involves designing dynamic pricing models, handling counter-proposals, and ensuring compliance through agent-to-agent communication.

Special Purpose AgentsHosted by Vera
Challenge brief

What you are building

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

To establish rules for AI web crawlers, this challenge tasks you with building an advanced A2A multi-agent system using LangGraph and Marvin to automate content licensing negotiations between 'Publisher Agents' and 'Crawler Agents.' Each agent type will be powered by Claude Opus 4.5, leveraging its sophisticated reasoning for contract understanding and negotiation. The system must implement graph-based workflows to manage negotiation states, enforce content rules (simulated), and ensure secure, auditable transactions via an MCP for contract finalization and compensation. This involves designing dynamic pricing models, handling counter-proposals, and ensuring compliance through agent-to-agent communication.

Datasets

Shared data for this challenge

Review public datasets and any private uploads tied to your build.

Loading datasets...
Learning goals

What you should walk away with

  • Master LangGraph for building complex, stateful Directed Acyclic Graph (DAG) workflows for multi-agent interactions.

  • Implement the A2A (Agent-to-Agent) protocol for secure, authenticated, and verifiable communication between distinct agent roles.

  • Design and create structured agent personas using Marvin, leveraging its declarative syntax for agent capabilities and output schemas.

  • Utilize Claude Opus 4.5 for its advanced natural language understanding, logical reasoning, and ability to process long-context legal-like documents for contract negotiation.

  • Develop extended thinking capabilities for agents to analyze counter-proposals, evaluate risks, and adapt negotiation tactics.

  • Build an MCP service for managing contract states, logging negotiation history, and integrating with simulated payment/blockchain systems for compensation.

  • Orchestrate a multi-agent simulation where Publisher Agents dynamically price content based on RSL 1.0 rules and Crawler Agents negotiate for access.

How this agent runs

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 strat...

Preview configuration

Challenge input

{'content_asset_id': 'abc123', 'publisher_rules': {'min_price_per_token': 0.001, 'licenses_available': ['commercial', 'research']}, 'crawler_intere...

Agent execution

The configured agent processes the input under the challenge policy.

Evaluated output

{'negotiation_id': '...', 'status': 'success'|'failure', 'final_agreement': {'content_asset_id': '...', 'license_type': 'commercial', 'price_per_to...

Checks for
  • Verifies that all signed agreements adhere to the initial RSL 1.0 rules (e.g., min price, license types).
  • Confirms that agents communicate securely and exclusively via the defined A2A protocol.
  • Checks if the final agreement is successfully recorded in the simulated MCP ledger.
Proof of success
  • NegotiationSuccessRate target: 80%
Runtime evidence
  • Python execution harness
View technical recipe

Configured tools

No tool records are attached.

Evaluation contract

  • The evaluation module defines the checks.

Recipe state

This is a preview. The configuration can change before the evaluation recipe is locked.

Run this agent on your dataset and AI stack

Bring your dataset, model providers, and success criteria. We will scope the right managed run for your team.

Scope a managed run
Start from your terminal
$npx -y @versalist/cli start ai-content-licensing-agent

[ok] Wrote CHALLENGE.md

[ok] Wrote .versalist.json

[ok] Wrote eval/examples.json

Requires VERSALIST_API_KEY. Works with any MCP-aware editor.

Docs
Manage API keys
Explore

Find another challenge

Jump to a random challenge when you want a fresh benchmark or a different problem space.

Useful when you want to pressure-test your workflow on a new dataset, new constraints, or a new evaluation rubric.

Frequently Asked Questions about AI Content Licensing Agent