Hugging Face's logo

Datasets:
Zovda
/
dm0012-vibe-coding-samples

Tasks:
Text Generation
Reinforcement Learning
Modalities:
Text
Formats:
json
Languages:
English
Size:
< 1K
Tags:
zovda
text
code
software-engineering
swe-agent
docker
Libraries:
Datasets
pandas
Polars
License:
Dataset card Data Studio Files
xet
 Community
dm0012-vibe-coding-samples / README.md
Zovda's picture
Zovda
Gated access + lead form fields + site link (UTM)
8a702f4 verified
preview code
|
Raw History Blame Contribute Delete
6.08 kB
metadata 
license: other
license_name: zovda-sample-terms
license_link: https://huggingface.co/datasets/Zovda/dm0012-vibe-coding-samples
language:
  - en
pretty_name: DM0012 Vibe Coding Repository-Level Task Samples / Vibe Coding 仓库级任务样例
size_categories:
  - n<1K
task_categories:
  - text-generation
  - reinforcement-learning
tags:
  - zovda
  - text
  - code
  - software-engineering
  - swe-agent
  - docker
  - agent
gated: manual
extra_gated_heading: Request access to this dataset sample
extra_gated_description: >-
  We review requests within 1 business day. Access is granted for evaluation and
  research use. Commercial licensing and larger cuts are handled separately by
  Zovda.


  Full dataset, custom cuts and pricing:
  https://www.zovda.com/lp/llm-training-data/code-agents/?utm_source=huggingface&utm_medium=gate_form&utm_campaign=dm0012
extra_gated_button_content: Request access
extra_gated_prompt: >-
  By requesting access you agree to use this sample for evaluation only, not to
  redistribute it, and that Zovda may contact you about the full dataset. We use
  your details only to respond to this request and to discuss licensing.
extra_gated_fields:
  Company: text
  Work email: text
  I want to use this dataset for:
    type: select
    options:
      - Fine-tuning
      - Pre-training
      - Evaluation / benchmark
      - Data research
      - Other
  Intended use:
    type: select
    options:
      - Commercial product
      - Academic research
      - Internal evaluation only
  Expected data volume:
    type: select
    options:
      - < 100 GB
      - 100 GB - 1 TB
      - '> 1 TB'
      - Not sure yet
  I agree to the sample terms: checkbox

DM0012 · Vibe Coding Repository-Level Task Samples / Vibe Coding 仓库级任务样例

Item / 项目 Value / 值
Dataset ID / 数据集编号 DM0012
Files / 数据文件 1
Size / 数据体积 29.36 KB
Provided by / 提供方 Zovda
Contact / 联系邮箱 lucas@zovda.com

Overview / 简介

Repository-level task corpus for coding-agent (SWE-Agent) training and evaluation: each package targets a real GitHub issue and bundles a runnable Docker sandbox, task description, reference patches and automatic grading scripts.

面向代码智能体（SWE-Agent）训练与评测的仓库级任务语料：每个任务包对应一个真实开源仓库的 GitHub Issue，含可运行 Docker 沙箱、任务描述、参考补丁与自动验收脚本。

What's inside / 数据内容

  • manifest.json: the public task-package manifest listing name, size, date, org, repo and issue id for each package.
  • manifest.json：随样例库公开的任务包清单，逐条记录包名、大小、日期、org、repo 与 issue 编号。
  • Full packages (<org>__<repo>-<issueId>.zip) are held in a private delivery store; about 42 entries per package — docker/<issueId>/ (OCI-layout sandbox image) plus json/<org__repo-issueId>/ (task description, code.patch, model.patch, final.diff, test.patch, check_delivery.py, grade.sh, dockerfile, setup_env.sh, setup_repo.sh, pin_by_date.py).
  • 完整任务包（<org>__<repo>-<issueId>.zip）存放于私有交付目录，每包约 42 个条目，结构为：docker/<issueId>/（OCI 布局沙箱镜像）与 json/<org__repo-issueId>/（任务描述、code.patch、model.patch、final.diff、test.patch、check_delivery.py、grade.sh、dockerfile、setup_env.sh、setup_repo.sh、pin_by_date.py）。
  • Contact lucas@zovda.com to obtain packages.
  • 取数请联系 lucas@zovda.com。

Specifications / 规格

  • Packaging: Docker (OCI layout) + JSON, two layers
  • 打包：Docker（OCI 布局）+ JSON 双层
  • Naming: <org>__<repo>-<issueId>, traceable to the real issue
  • 命名：<org>__<repo>-<issueId>，issueId 可回溯真实 Issue
  • Acceptance: grade.sh auto-grades inside the sandbox; check_delivery.py verifies delivery completeness
  • 验收：grade.sh 在沙箱内自动评分，check_delivery.py 校验交付完整性

Suitable tasks / 适用任务

  • Code-model post-training (SFT): paired "issue description → reference patch" samples
  • 代码模型后训练（SFT）：「Issue 描述 → 参考补丁」成对样本
  • Reinforcement learning / agent training: sandbox + auto-grading provide reward signals
  • 强化学习 / Agent 训练：沙箱 + 自动评分提供奖励信号
  • End-to-end coding-agent evaluation: issue-resolution pass rate
  • 代码智能体端到端评测：Issue 修复通过率

Files / 文件清单

# File Size
1 vibeCoding/manifest.json 29.36 KB

Usage / 使用方式 

from huggingface_hub import snapshot_download

snapshot_download(
    repo_id="Zovda/dm0012-vibe-coding-samples",
    repo_type="dataset",
    local_dir="./dm0012-vibe-coding-samples",
)

Rights and disclaimer / 授权与声明

Tasks are built from real issues in public open-source repositories, which retain their original open-source licences. This manifest demonstrates task organisation; packages are licensed on request.

任务基于公开开源仓库的真实 Issue 构建，仓库代码遵循其原始开源许可。本清单用于展示任务组织方式，任务包按需授权交付。

This dataset is shared to demonstrate data organisation, technical specifications and delivery structure. For formal licensing, large-scale access or custom production, contact lucas@zovda.com.

本数据集仅用于展示数据组织方式、技术规格与交付结构。正式授权、规模化取数与定制生产请联系 lucas@zovda.com。

About Zovda / 关于 Zovda

Zovda builds, cleans, annotates and packages training data across video, audio, images, e-commerce image-text pairs, exam question banks, embodied AI and coding-agent corpora.

Zovda 提供多模态与文本训练语料的数据采集、清洗、标注与结构化交付服务，覆盖视频、音频、图像、电商图文、K12 与职业题库、具身智能、代码智能体等方向。

Contact / 联系方式: lucas@zovda.com