nash-env

What problem does it solve?

Help users identify which existing Nobel game-theory environment best matches a described real-world situation, so they can run the correct simulations and equilibrium checks.

Core Features & Use Cases

• Problem-to-model classification: Recommends the most suitable game environment (e.g., Hawk-Dove, Repeated Prisoner’s Dilemma, Public Goods) after confirming key assumptions like goals, information structure, and time horizon.
• Agent-team parallel reasoning: When the match is unclear or cross-domain, coordinates multiple subagents to compare candidate models and produce a scored comparison table.
• Source-code guided environment understanding: Explains environment mechanics and points to the relevant environment implementation files so users can inspect how payoffs and equilibria are computed.
• Memory-guided routing: Persists the user’s final model selection and routes the workflow to the simulation execution skill.

Quick Start

Ask the AI: "Given this scenario [describe participants, incentives, information, and whether interactions repeat], which Nobel game-theory model fits best, and what assumptions do you need from me?"