name: bazaarbot
version: "1.0.0"
description: >
  Customer-vendor price negotiation environment with asymmetric information,
  career dynamics, and Rubinstein bargaining theory. Simulates real-world
  informal market negotiation (kirana stores, vendor stalls, service pricing).

observation:
  type: object
  model: server.models.BazaarObservation
  fields:
    current_round: { type: integer, description: "Current negotiation round" }
    max_rounds: { type: integer, description: "Maximum rounds in this episode" }
    own_last_offer: { type: number, nullable: true, description: "Buyer's last offer" }
    opponent_last_offer: { type: number, nullable: true, description: "Seller's current ask" }
    own_private_budget: { type: number, description: "Buyer's private budget (hidden from seller)" }
    rounds_remaining: { type: integer, description: "Rounds left in episode" }
    seller_last_move_delta: { type: number, nullable: true, description: "How much seller conceded last round" }
    item_name: { type: string, description: "Item being negotiated" }
    seller_asking_price: { type: number, description: "Seller's opening anchor price" }
    episode_number: { type: integer, description: "Current episode in career" }
    total_episodes: { type: integer, description: "Total episodes in task" }
    message: { type: string, description: "Seller's spoken message" }

action:
  type: object
  model: server.models.BazaarAction
  fields:
    action: { type: string, enum: ["offer", "accept", "walk"], description: "Action type" }
    price: { type: number, nullable: true, description: "Offer price (required for 'offer' action)" }

reward:
  type: object
  model: server.models.BazaarReward
  range: [0.0, 1.0]
  partial_progress: true
  fields:
    reward: { type: number, description: "Step reward" }
    terminal: { type: boolean, description: "Whether episode ended" }
    components: { type: object, description: "Reward breakdown" }

tasks:
  single_deal:
    difficulty: easy
    description: "Single negotiation. Symmetric information. Moderate seller."
    max_steps: 8
    grader: "server.tasks.grade_single_deal"
    success_threshold: 0.3

  asymmetric_pressure:
    difficulty: medium
    description: "Hidden deadline + inventory pressure. Infer seller urgency."
    max_steps: 8
    grader: "server.tasks.grade_asymmetric_pressure"
    success_threshold: 0.4

  career_10:
    difficulty: hard
    description: "10 consecutive deals. Seller adapts to buyer patterns."
    max_steps: 80
    grader: "server.tasks.grade_career_10"
    success_threshold: 0.5

endpoints:
  reset: { method: POST, path: /reset }
  step: { method: POST, path: /step }
  state: { method: GET, path: /state }
  score: { method: GET, path: /score }
  health: { method: GET, path: /health }
  tasks: { method: GET, path: /tasks }