Generated by Repo2RLEnv, this dataset turns real GitHub repositories into verifiable reinforcement learning environments. The dataset includes specifications, instructions, oracle patches, test scripts, and Dockerfiles for tasks derived from 13 source repositories, including popular projects like Flask, Requests, and Gin. It was created by author AdithyaSK and last updated on May 27, 2026.
Use Cases
- Training RL agents for automated code patching based on provided oracle patches.
- Benchmarking code generation models using the structured task instructions and test scripts.
- Developing environments for testing software robustness using the included Docker configurations.
Strengths
- Derived from 13 established open-source projects, providing a diverse set of real-world codebases.
- Includes multiple components per task: spec, instruction, oracle patch, test script, and Dockerfile for verifiable environments.
Limitations
- Column-level documentation is absent; field semantics must be inferred after download.
- Row count is unknown, which may limit suitability assessment.
- Description metadata is limited; actual data quality requires manual inspection after download.
Provenance
- Source
- 13 GitHub repositories including encode/httpx, gin-gonic/gin, pallets/flask, psf/requests.
- Collection Method
- Generated by the Repo2RLEnv tool.
- Freshness
- Last updated 2026-05-27 08:06:38.