name: repair-strategy-system-v2
version: "2.0.0"
description: >
  A deterministic environment that simulates a distributed system with hidden state.
  Agents must infer root causes from metrics and ambiguous logs.
  Enforces cascading failures, delayed recovery, and action costs.

environment:
  module: env_v2
  class: RepairEnvV2
  seed: 42

tasks:
  - id: easy
    description: "Database failure with clear symptoms. Single root cause."
    max_steps: 10
    optimal_steps: 3
  - id: medium
    description: "Cache failure and passive queue degradation. Misleading symptoms."
    max_steps: 15
    optimal_steps: 5
  - id: hard
    description: "Full cascading failure starting from DB. API and Queue both impacted."
    max_steps: 20
    optimal_steps: 7
  - id: multi_root
    description: "Independent failures in both DB and Queue. High complexity."
    max_steps: 25
    optimal_steps: 10

action_space:
  type: structured
  model: models_v2.Action
  fields:
    action_type:
      type: enum
      values:
        - restart_service
        - scale_up
        - clear_cache
        - repair_database
        - inspect_database
        - inspect_api
        - inspect_cache
        - inspect_queue
        - no_op
    target:
      type: enum
      values:
        - database
        - api
        - cache
        - queue

observation_space:
  type: structured
  model: models_v2.SymptomObservation
  fields:
    metrics:
      type: dict
      description: "Map of component name to its metrics (latency, error_rate, etc.)"
    logs:
      type: list
      description: "List of highly ambiguous log entries from various components"
    step_count:
      type: int
    max_steps:
      type: int
    inspections_remaining:
      type: int
      description: "Counter for free inspections (first 3 are free)"

reward:
  model: models_v2.Reward
  range: [-5.0, 10.0]
  description: >
    Dense reward: +1.5 for root cause repair started, +0.5 for symptom improvement,
    -1.0 for wrong component targeted, +5.0 terminal bonus, -0.2 step penalty.
    Includes Action Costs (V3): e.g., -0.5 for DB/Scale, -0.2 for Restart/Cache.

graders:
  module: graders_v2
  function: grade_v2

baseline:
  module: baseline_v2
  function: run_task

server:
  host: "0.0.0.0"
  port: 7860
  endpoints:
    - path: /tasks
      method: GET
    - path: /grader
      method: POST
    - path: /baseline
      method: GET
    - path: /diagnose
      method: POST
      description: "Submit a theoretical diagnosis for immediate feedback"
    - path: /reset
      method: POST
    - path: /step
      method: POST
    - path: /state
      method: GET