vedkdev/FlakyTestSleuthOpenEnvRL

1spec_version: 12name: flaky_sleuth3type: space4runtime: fastapi5app: server.app:app6port: 80007 8version: 0.1.09description: >10  An RL environment where an LLM agent investigates flaky tests in Python repositories.11  The agent uses tool-like actions to read files, search code, and run tests, then submits12  a terminal verdict for classification, root-cause detection, or fix proposal.13 14action_type: FlakySleuthAction15observation_type: FlakySleuthObservation16reward_range: (0.001, 0.999)17episode_max_steps: 2018baseline_script: inference.py19 20tasks:21  - id: task1_classify22    name: Flaky vs Stable Classification23    difficulty: easy24    description: Classify the target test as flaky or stable.25  - id: task2_root_cause26    name: Root Cause Category Identification27    difficulty: medium28    description: Predict flaky-test root-cause category (OD, NOD, TD, TZD, NIO, ID, etc.).29  - id: task3_fix_proposal30    name: Fix Proposal31    difficulty: hard32    description: Propose a concrete fix as unified diff for a known flaky test.33 34infra:35  vcpu: 236  memory_gb: 837  max_inference_minutes: 2038