Dataset Viewer (First 5GB)
Auto-converted to Parquet Duplicate
question_category
stringclasses
22 values
complexity_level
stringclasses
3 values
question
stringlengths
58
5.99k
⌀
agent_prompt
stringlengths
1.14k
3.65k
⌀
enabled_tools
listlengths
0
4
skills_path
stringclasses
1 value
uuid
stringlengths
36
36
messages
listlengths
3
173
tools
listlengths
5
5
metadata
dict
hf_split
stringclasses
1 value
fact_seeking_qa
intermediate
What are the key differences between using `async/await` and `.then()` for handling promises in JavaScript, and when should I prefer one approach over the other in terms of readability and error handling?
# Async/Await vs .then() Promise Handling Guide ## Problem Summary Compare JavaScript promise handling approaches (`async/await` vs `.then()`) focusing on readability and error handling scenarios. ## Requirements - JavaScript ES2017+ for async/await syntax - Understanding of Promise fundamentals - Knowledge of synchr...
[ "bash", "todoread", "todowrite", "skill" ]
/nemo_run/code/.skills/bash_skills
ae216f9f-ec63-56ad-8697-5868d07ca94d
[ { "role": "system", "content": "You are opencode, an interactive CLI tool that helps users with software engineering tasks. Use the instructions below and the tools available to you to assist the user.\n\nIMPORTANT: Refuse to write code or explain code that may be used maliciously; even if the user claims i...
[ { "id": "invalid", "description": "Do not use", "inputSchema": { "jsonSchema": { "$schema": "https://json-schema.org/draft/2020-12/schema", "type": "object", "properties": { "tool": { "type": "string" }, "error": { "type...
{ "uuid": "e6dae5c1-a469-505c-97ad-227569125a2d" }
bash_only_tool_skills
reasoning
advanced
I'm building a distributed system where multiple services need to coordinate state changes using event sourcing. I'm trying to decide between implementing a saga pattern with orchestration versus choreography for handling cross-service transactions. What are the key trade-offs in terms of consistency guarantees, failur...
# Saga Pattern Decision Guide: Orchestration vs Choreography ## Problem Summary Choosing between orchestrated and choreographed sagas for distributed fund transfers across banking services, requiring atomic commit/rollback behavior with proper failure recovery. ## Core Trade-offs **Orchestration:** - Centralized coo...
[ "bash", "todoread", "todowrite", "skill" ]
/nemo_run/code/.skills/bash_skills
82d805af-9b0e-59f7-9cc8-316a4e51b83a
[ { "role": "system", "content": "You are opencode, an interactive CLI tool that helps users with software engineering tasks. Use the instructions below and the tools available to you to assist the user.\n\nIMPORTANT: Refuse to write code or explain code that may be used maliciously; even if the user claims i...
[ { "id": "invalid", "description": "Do not use", "inputSchema": { "jsonSchema": { "$schema": "https://json-schema.org/draft/2020-12/schema", "type": "object", "properties": { "tool": { "type": "string" }, "error": { "type...
{ "uuid": "53cc209e-de8e-5425-9514-696041e43568" }
bash_only_tool_skills
code_explanation
advanced
I'm trying to understand how the Redux Toolkit's createAsyncThunk function works under the hood, particularly the middleware chain and how it handles pending/rejected/fulfilled states. Can you walk me through the implementation details of how it dispatches actions and manages the promise lifecycle, especially in relati...
# Redux Toolkit createAsyncThunk Implementation Guide ## Problem Summary Understanding the internal mechanics of Redux Toolkit's `createAsyncThunk`, focusing on middleware integration, promise lifecycle management, and action dispatching within Redux's architecture. ## Core Requirements - Knowledge of Redux middlewar...
[ "bash", "todoread", "todowrite", "skill" ]
/nemo_run/code/.skills/bash_skills
544ad8b8-fc80-5542-af2b-8df03c34896a
[ { "role": "system", "content": "You are opencode, an interactive CLI tool that helps users with software engineering tasks. Use the instructions below and the tools available to you to assist the user.\n\nIMPORTANT: Refuse to write code or explain code that may be used maliciously; even if the user claims i...
[ { "id": "invalid", "description": "Do not use", "inputSchema": { "jsonSchema": { "$schema": "https://json-schema.org/draft/2020-12/schema", "type": "object", "properties": { "tool": { "type": "string" }, "error": { "type...
{ "uuid": "629a13b4-90d9-5dae-a3d2-55b7e747e8e8" }
bash_only_tool_skills
reasoning
beginner
I'm building a simple calculator app and need to decide between using integers or floating-point numbers for all calculations. What are the trade-offs of each approach, and which would you recommend for a beginner-friendly calculator that handles basic arithmetic operations?
# Calculator Data Type Selection Guide ## Problem Summary Choosing between integers and floating-point numbers for a beginner calculator app that performs basic arithmetic operations. ## Requirements - Support addition, subtraction, multiplication, and division - Handle decimal inputs and results - Maintain simplicit...
[ "bash", "todoread", "todowrite", "skill" ]
/nemo_run/code/.skills/bash_skills
56a5152f-9653-5aa4-b3e7-61d593c555f4
[ { "role": "system", "content": "You are opencode, an interactive CLI tool that helps users with software engineering tasks. Use the instructions below and the tools available to you to assist the user.\n\nIMPORTANT: Refuse to write code or explain code that may be used maliciously; even if the user claims i...
[ { "id": "invalid", "description": "Do not use", "inputSchema": { "jsonSchema": { "$schema": "https://json-schema.org/draft/2020-12/schema", "type": "object", "properties": { "tool": { "type": "string" }, "error": { "type...
{ "uuid": "b468c5f5-955c-5d4c-8122-2b540f1f7777" }
bash_only_tool_skills
fact_seeking_qa
beginner
"What is the correct syntax for creating a new React component using hooks, and how do I properly in(...TRUNCATED)
"# Creating React Components with Hooks - Beginner Guide\n\n## Problem Summary\nCreating functional (...TRUNCATED)
[ "bash", "todoread", "todowrite", "skill" ]
/nemo_run/code/.skills/bash_skills
98f8b11f-87f2-5176-8a60-f472cab2ae07
[{"role":"system","content":"You are opencode, an interactive CLI tool that helps users with softwar(...TRUNCATED)
[{"id":"invalid","description":"Do not use","inputSchema":{"jsonSchema":{"$schema":"https://json-sch(...TRUNCATED)
{ "uuid": "cb33147a-5bb2-5377-8bef-1bacc8e5fb1b" }
bash_only_tool_skills
brainstorming
advanced
"I'm designing a distributed system for real-time collaborative document editing that needs to handl(...TRUNCATED)
"# Distributed Real-Time Collaborative Editing Architecture Guide\n\n## Problem Summary\nDesign a di(...TRUNCATED)
[ "bash", "todoread", "todowrite", "skill" ]
/nemo_run/code/.skills/bash_skills
8bb49ff3-35b0-592e-96c3-514887c73379
[{"role":"system","content":"You are opencode, an interactive CLI tool that helps users with softwar(...TRUNCATED)
[{"id":"invalid","description":"Do not use","inputSchema":{"jsonSchema":{"$schema":"https://json-sch(...TRUNCATED)
{ "uuid": "2ab2321c-1ac9-5157-a21c-a024045dbc15" }
bash_only_tool_skills
planning_and_task_structuring
advanced
"I'm building a microservices architecture for a real-time analytics platform that processes streami(...TRUNCATED)
"# Microservices Architecture Planning Guide\n\n## Problem Summary\nDesign a horizontally scalable, (...TRUNCATED)
[ "bash", "todoread", "todowrite", "skill" ]
/nemo_run/code/.skills/bash_skills
53d86d44-bd3a-5b7f-b7ae-296f4c814459
[{"role":"system","content":"You are opencode, an interactive CLI tool that helps users with softwar(...TRUNCATED)
[{"id":"invalid","description":"Do not use","inputSchema":{"jsonSchema":{"$schema":"https://json-sch(...TRUNCATED)
{ "uuid": "d3b607aa-95e3-5f22-b7df-b078c098ce59" }
bash_only_tool_skills
code_explanation
beginner
"I'm learning Python and came across this code snippet that uses list comprehension with a condition(...TRUNCATED)
"## Problem Summary\n\nThis is a code explanation request about Python list comprehension with condi(...TRUNCATED)
[ "bash", "todoread", "todowrite", "skill" ]
/nemo_run/code/.skills/bash_skills
12d6a08a-85ea-51a0-a314-d783ab65f698
[{"role":"system","content":"You are opencode, an interactive CLI tool that helps users with softwar(...TRUNCATED)
[{"id":"invalid","description":"Do not use","inputSchema":{"jsonSchema":{"$schema":"https://json-sch(...TRUNCATED)
{ "uuid": "28a92724-be8d-5d96-be6d-a0455d1f2bd2" }
bash_only_tool_skills
code_generation
beginner
"I'm building a simple calculator app and need help creating a function that takes two numbers and a(...TRUNCATED)
"# Simple Calculator Function Guide\n\n## Problem Summary\nCreate a Python function that performs ba(...TRUNCATED)
[ "bash", "todoread", "todowrite", "skill" ]
/nemo_run/code/.skills/bash_skills
131c3ad1-28d5-5049-bb61-876997cdb266
[{"role":"system","content":"You are opencode, an interactive CLI tool that helps users with softwar(...TRUNCATED)
[{"id":"invalid","description":"Do not use","inputSchema":{"jsonSchema":{"$schema":"https://json-sch(...TRUNCATED)
{ "uuid": "f88b7136-48a8-59e9-9faa-070e9bdeacf8" }
bash_only_tool_skills
code_explanation
intermediate
"Can you explain how this React custom hook for managing form state works? I'm particularly confused(...TRUNCATED)
"## Problem Summary\n\nThis React custom hook manages form state using `useReducer` and exposes cont(...TRUNCATED)
[ "bash", "todoread", "todowrite", "skill" ]
/nemo_run/code/.skills/bash_skills
96c8bc77-3fed-53cd-beb7-1a0356047fc3
[{"role":"system","content":"You are opencode, an interactive CLI tool that helps users with softwar(...TRUNCATED)
[{"id":"invalid","description":"Do not use","inputSchema":{"jsonSchema":{"$schema":"https://json-sch(...TRUNCATED)
{ "uuid": "c04225f5-3a19-5971-b779-62c5d0682cf3" }
bash_only_tool_skills
End of preview. Expand in Data Studio

nvidia/Nemotron-SFT-OpenCode-v1(複製)

データの内容

LLM が OpenCode の CLI の中で動く能力を高めるための、指示チューニング用のデータ。サブセットは general、bash_only_tool、agent_skills、question_tool など 6 種類で、ツールの組み合わせや AGENTS.md の有無で分けてある。質問とエージェントのスキルは GPT OSS 120B で合成し、解答の軌跡は Qwen3-Coder-480B-A35B-Instruct で生成した。ライセンスは CC BY 4.0。

データの概要

項目 値
行数・列数 不明(データセットビューアで読めない形式。ファイルの内訳だけを示す)
容量 32.69 GB(ファイル 9 本)
ファイルの形式 jsonl 6、json 1
言語 en
公開 public

全データセットの一覧は HYUKI5/dataset-catalog にある。

複製について

nvidia/Nemotron-SFT-OpenCode-v1 をリビジョン 556d5237acff203f3e1a0be49428634c3606cda2 の時点でそのまま複製したもの。中身は変えていない。判断モデル Laya(convaiinnovations/laya)の微調整に使う候補として、調査メモ(Laya 学習用データセット調査)に載せたデータを集めている。

出典

論文

  • 元のカードに論文へのリンクは見つからなかった

アノテーションの方法

データセットの作成日:

01/09/2026

微調整に使うときの注意

  • 調査メモの台帳での区分: A(許諾型のライセンスだけで、疑義の注記が無い)。区分は記載の文言で機械的に付けたもので、法的な判定ではない。学習に入れる前に元のカードとライセンスの原文で確かめる。
  • Laya の入力は 2,048 token まで(ModernBERT-large / mmBERT)。長い本文は先頭から黙って切られるので、切れる行は事前に除くか、要約・切り出しをしてから入れる。
  • 分割は記事・文書単位で行い、同じ本文から作った別の問いを学習側と較正側に分けない。温度と閾値は学習に使っていない分割で別々に当てる。

調査メモの記述

メモ 節 ライセンス(記載) 確認 注記
nvidia_nemotron.md 付録. 取得した nvidia 組織のデータセット全件の簡易一覧 (L269) cc-by-4.0 確認済 確認印なし(各ファイルの凡例により確認済扱い) / 原文注記: × コード・パズル・化学

元のカード(日本語訳)

以下は元のデータセットのカードの本文を日本語に訳したもの(機械翻訳)。訳に疑問があるときは、末尾の原文を確かめる。


データセットの説明:

Nemotron-SFT-OpenCode-v1 は、エージェント向けの指示チューニング用データセットであり、大規模言語モデル(LLM)が OpenCode のコマンドラインインターフェース(CLI)フレームワークの中で動作する能力を高め、ツール呼び出しやエージェントのスキルといった単純な能力を身につけさせる。

このデータセットは商用・非商用のどちらでも利用できる。

データセットのサブセット:

Nemotron-SFT-OpenCode-v1 には次のサブセットがある。質問とエージェントのスキルは GPT OSS 120B モデルで合成的に生成され、エージェントの解答の軌跡は Qwen3-Coder-480B-A35B-Instruct で生成された。

  • general

エージェント型 CLI によく尋ねられる一般的な質問を含むサブセット。各質問は 2 組ずつある。一方は AGENTS.md を付けないもの、もう一方は同じ質問に独自の AGENTS.md を添えたものである。

  • bash_only_tool

許可されるツールを todoread/write と bash だけに絞ったサブセット。システムプロンプトには、元の OpenCode のすべてのツールを bash コマンドで対応づけた、詳しい AGENTS.md が追加される。これは、事前学習済みモデルに本来備わっていない機能でも、その存在を把握させ、同等のコマンドを幅広く使うよう促すためである。

  • agent_skills

元の質問に、質問そのものを合成したのと同じ分類体系を使って作った、合成のエージェントのスキルがいくつか付いたサブセット。モデルはこれを読んで、その質問に関連するスキルを探し、動的に読み込める。同じカテゴリを改善するスキルをまとめられるようにもなっている。

  • bash_only_tool_skills

ツールの集合を todoread/write、bash、skill だけに絞ったサブセット。元の bash の長いプロンプトの代わりに、OpenCode の各ツールをそれぞれ 1 つのスキルとして記述したエージェントのスキルの集合を使う。各スキルは、元のツールを実装する bash コマンドを記述する。そのためモデルは、必要なスキルだけを読むことができる。

  • question_tool

question ツールを有効にし、ロールアウト中にモデルが「ユーザー」に質問できるようにしたサブセット。ユーザーの返答は、ロールアウトを生成したのと同じモデルに専用のプロンプトを与えて合成している。そのため、機能を実装する前に明確化が必要になる状況でも、推論時に頑健であるよう促される。

  • agent_skills_question_tool

エージェントのスキルと、ユーザーに質問する能力を組み合わせたサブセット。モデルは、コンテキストに読み込まれたスキルを探して区別する前に、ユーザーの意図を明確にできる。

データセットの提供者:

NVIDIA Corporation

データセットの作成日:

01/09/2026

ライセンス・利用条件:

適用される規約: このデータセットは Creative Commons Attribution 4.0 International License(CC BY 4.0) のもとで提供される。

想定される利用:

このデータセットは、実世界の issue の記述を解釈し、複数段階のツール利用を計画し、コードベースを探索して、SWE-Bench 型の設定で修正を実装しなければならない、自律的なソフトウェア工学エージェントやコード志向のアシスタントを作る LLM エンジニアと研究チームを想定している。SFT や、そのようなモデルの蒸留に向いている。軌跡は、エージェントの方策の評価とデバッグ、リポジトリを踏まえた推論の改善、回帰を起こさないコード編集の振る舞いの研究にも使える。学術と実運用の両方の環境が対象となる。

データセットの特性

** データ収集方法
ハイブリッド: 自動, 合成

** ラベル付けの方法
ハイブリッド: 自動, 合成

データの形式

Modality: テキスト Format: JSONL Structure: テキスト+メタデータ

データ量の集計

Simplified Name Sample Count
general 90K
bash_only_tool 97K
bash_only_tool_skills 96K
question_tool 76K
agent_skills 67K
agent_skills_question_tool 33K
合計 459K

データの総容量: 約 30 GB

参考文献:

OpenCode

倫理的配慮:

NVIDIA は、信頼できる AI は共同の責任であると考えており、多様な AI 応用の開発を可能にする方針と慣行を整えている。ダウンロードして利用規約に従って使う場合、開発者は社内の開発チームと協力して、このデータセットが関連する業界と用途の要件を満たすこと、そして想定外の製品上の誤用に対処することを確かめるべきである。 品質、リスク、セキュリティの脆弱性、または NVIDIA の AI に関する懸念は こちら から報告してほしい。

原文(英語)

Dataset Description:

Nemotron-SFT-OpenCode-v1 is an agentic instruction tuning dataset that enhances the ability of Large Language Models (LLMs) to operate within the OpenCode Command Line Interface (CLI) framework and instills simple capabilities such as tool calling and agent skills.

This dataset is ready for commercial/non-commercial use.

Dataset Subsets:

Nemotron-SFT-OpenCode-v1 contains the following subsets, where the questions and agent skills are synthetically generated using the GPT OSS 120B model and whose agentic solution traces were generated using Qwen3-Coder-480B-A35B-Instruct.

  • general

Subset which contains general questions that are commonly asked to Agentic CLI. Each question comes in two pairs - one without an accompanying AGENTS.md and the same question accompanied by a custom AGENTS.md.

  • bash_only_tool

Subset where the list of allowed tools is limited to just todoread/write and bash. The system prompt is augmented with a detailed AGENTS.md which describes the equivalent of all the original OpenCode tools in terms of bash commands and encourages broadly equivalent commands so that model is aware of capabilities even if not inherent to the pretrained model.

  • agent_skills

Subset where the original question is accompanied by a list of several synthetic agent skills constructed using the same taxonomy used to synthesize the questions themselves. This enables the model to scan and dynamically load skills that are relevant to the particular question, and allows for the grouping of skills that broadly improve the same category.

  • bash_only_tool_skills

Subset which restricts tool set to just todoread/write, bash and skill, and replaces the large bash prompt in the original subset with a set of agent skills that describe each individual opencode tool as a skill - each skill then describes the bash commands which implement the original tool. As such, the model can opt to read just the skills it needs.

  • question_tool

Subset where the question tool has been enabled in order to allow the model to ask the “user” questions during rollout. The user’s feedback is synthesized using a custom prompt with the same model that generates the rollout. This encourages the model to be robust at inference time to situations where clarifications may be needed before implementing a feature.

  • agent_skills_question_tool

Subset where the agent skills are combined with the ability to ask user questions, enabling the model to clarify the user’s intent before seeking information from and disambiguating between skills that are loaded into context.

Dataset Owner(s):

NVIDIA Corporation

Dataset Creation Date:

01/09/2026

License/Terms of Use:

GOVERNING TERMS: This dataset is governed by the Creative Commons Attribution 4.0 International License (CC BY 4.0)

Intended Usage:

This dataset is intended for LLM engineers and research teams building autonomous software engineering agents and code-focused assistants. It is suitable for supervised fine-tuning and distillation of models that must interpret real-world issue statements, plan multi-step tool use, navigate codebases, and implement fixes in a SWE-Bench–style setting. The trajectories can also be used to benchmark and debug agent policies, improve repository-aware reasoning, and study robust, regression-free code editing behaviors in both academic and production environments.

Dataset Characterization

** Data Collection Method
Hybrid: Automated, Synthetic

** Labeling Method
Hybrid: Automated, Synthetic

Dataset Format

Modality: Text Format: JSONL Structure: Text + Metadata

Dataset Quantification

Simplified Name Sample Count
general 90K
bash_only_tool 97K
bash_only_tool_skills 96K
question_tool 76K
agent_skills 67K
agent_skills_question_tool 33K
Total 459K

Total Data Storage: ~30 GB

Reference(s):

OpenCode

Ethical Considerations:

NVIDIA believes Trustworthy AI is a shared responsibility and we have established policies and practices to enable development for a wide array of AI applications. When downloaded or used in accordance with our terms of service, developers should work with their internal developer teams to ensure this dataset meets requirements for the relevant industry and use case and addresses unforeseen product misuse.
Please report quality, risk, security vulnerabilities or NVIDIA AI Concerns here.

Downloads last month
74