Tool tree

marketbench.vercel.app

AI & AgentsTools verifiedOct 26, 2026

A reproducible WebMCP benchmark for browser agents making economic decisions.

get_benchmark_rulesAnswer

Read the locked benchmark objective, limits, active scenario, seed and week.

View tool JSON

{
  "name": "get_benchmark_rules",
  "kind": "answer",
  "impl": "imperative",
  "description": "Read the locked benchmark objective, limits, active scenario, seed and week.",
  "inputSchema": {
    "type": "object",
    "properties": {},
    "additionalProperties": false
  }
}

get_market_stateAnswer

Read only market information available to a manager at the start of the current week.

View tool JSON

{
  "name": "get_market_state",
  "kind": "answer",
  "impl": "imperative",
  "description": "Read only market information available to a manager at the start of the current week.",
  "inputSchema": {
    "type": "object",
    "properties": {},
    "additionalProperties": false
  }
}

get_sales_historyAnswer

Read a compact window of prior weekly sales.

View tool JSON

{
  "name": "get_sales_history",
  "kind": "answer",
  "impl": "imperative",
  "description": "Read a compact window of prior weekly sales.",
  "inputSchema": {
    "type": "object",
    "properties": {
      "window": {
        "type": "integer",
        "minimum": 1,
        "maximum": 4
      }
    },
    "required": [
      "window"
    ],
    "additionalProperties": false
  }
}

set_pricesAction

Stage exactly one complete price update for the current week.

View tool JSON

{
  "name": "set_prices",
  "kind": "act",
  "impl": "imperative",
  "description": "Stage exactly one complete price update for the current week.",
  "inputSchema": {
    "type": "object",
    "properties": {
      "prices": {
        "type": "array",
        "minItems": 3,
        "maxItems": 3,
        "items": {
          "type": "object",
          "properties": {
            "product_id": {
              "type": "string",
              "enum": [
                "water",
                "cola",
                "chips"
              ]
            },
            "price": {
              "type": "number",
              "minimum": 0.8,
              "maximum": 5
            }
          },
          "required": [
            "product_id",
            "price"
          ],
          "additionalProperties": false
        }
      },
      "rationale": {
        "type": "string",
        "maxLength": 240
      }
    },
    "required": [
      "prices",
      "rationale"
    ],
    "additionalProperties": false
  }
}

advance_weekAction

Apply staged prices and simulate exactly one week. Rejects a stale expected_week.

View tool JSON

{
  "name": "advance_week",
  "kind": "act",
  "impl": "imperative",
  "description": "Apply staged prices and simulate exactly one week. Rejects a stale expected_week.",
  "inputSchema": {
    "type": "object",
    "properties": {
      "expected_week": {
        "type": "integer",
        "minimum": 1,
        "maximum": 12
      }
    },
    "required": [
      "expected_week"
    ],
    "additionalProperties": false
  }
}

finish_runAction

Close a 12-week run and return measured totals plus the calibrated 0-100 score.

View tool JSON

{
  "name": "finish_run",
  "kind": "act",
  "impl": "imperative",
  "description": "Close a 12-week run and return measured totals plus the calibrated 0-100 score.",
  "inputSchema": {
    "type": "object",
    "properties": {
      "expected_week": {
        "type": "integer",
        "const": 13
      }
    },
    "required": [
      "expected_week"
    ],
    "additionalProperties": false
  }
}