File size: 1,409 Bytes
4817a39
e7e76f5
717bee1
 
4817a39
 
 
 
 
717bee1
 
 
 
e7e76f5
 
 
 
717bee1
 
 
 
e7e76f5
 
 
 
 
717bee1
 
 
 
e7e76f5
 
 
 
 
717bee1
 
 
e7e76f5
717bee1
 
e7e76f5
 
717bee1
e7e76f5
717bee1
e7e76f5
717bee1
e7e76f5
717bee1
e7e76f5
717bee1
 
e7e76f5
717bee1
 
e7e76f5
717bee1
e7e76f5
717bee1
e7e76f5
717bee1
e7e76f5
717bee1
e7e76f5
717bee1
e7e76f5
717bee1
e7e76f5
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
name: "Quant-Gym"
description: "Financial analysis environment for testing AI agents"
version: "1.0.0"

environment:
  name: "QuantGym"
  class: "TradingEnvironment"
  module: "server.environment"

tasks:
  - id: "task1"
    name: "Fetch Market Data"
    description: "Agent must retrieve current price for AAPL"
    difficulty: "easy"
    grader:
      type: deterministic
      endpoint: /grader
    max_score: 1.0
    
  - id: "task2"
    name: "News Sentiment Analysis"
    description: "Agent must analyze news and recommend action"
    difficulty: "medium"
    grader:
      type: llm_judge
      endpoint: /grader
    max_score: 1.0
    
  - id: "task3"
    name: "Backtest Strategy"
    description: "Agent must backtest a trading strategy"
    difficulty: "hard"
    grader:
      type: deterministic
      endpoint: /grader
    max_score: 1.0

action_schema:
  type: object
  properties:
    type:
      type: string
      enum: [GET_PRICE, GET_NEWS, BUY, SELL, BACKTEST]
    symbol:
      type: string
    amount:
      type: integer
    explanation:
      type: string
    strategy:
      type: string

observation_schema:
  type: object
  properties:
    timestamp:
      type: string
    price:
      type: number
    balance:
      type: number
    holdings:
      type: integer
    portfolio_value:
      type: number
    last_news:
      type: object
    backtest_results:
      type: object