一手來源技術研究

一手來源技術研究

優先閱讀官方文件、程式碼與規格,建立附引用的技術研究。

Community · 社群來源 · 研究

來源狀態

可用來源

原始名稱: research

原始作者: PracticalSwan

這是第三方 Community Skill;2lus 收錄不代表已完整安全審計或保證安全。

資源類型

Skill

與類似 Skill 有什麼不同?

  • OpenAI 官方文件研究

    OpenAI Docs 聚焦 OpenAI 產品文件;Primary Source Research 跨來源查證與整理證據,不侷限特定供應商。

  • 技術寫作

    Technical Writing 將工程脈絡寫成 README、API 文件與 ADR;Primary Source Research 先查證資料與證據可信度。

來源描述的能力

這些是來源描述的可能操作,不代表 2lus 已授予權限或已測試。

  • 讀取檔案
  • 寫入檔案
  • 網路存取
原始來源 ↗

這個 Skill 是什麼?

將技術問題拆成可查證主張,保留來源、版本與不確定性,讓架構決策可回溯而非只提供無來源結論。

可以做什麼?何時適合使用?

  • 確認 Java library 的目前 API。
  • 比較相近函式庫的架構限制。
  • 釐清文件與實作行為的差異。

如何使用

  1. 列出研究問題、版本範圍與決策需要。
  2. 查官方文件、repository、規格和 release notes,記錄支持與衝突證據。
  3. 整理有引用的結論,分開已知、推論與尚待實測項目。

使用前你需要準備

  • 研究問題與讀者
  • 可公開來源與版本範圍

你可以替換:

[topic]、[version]

環境與相依需求

  • 可存取一手來源的網路及保存研究筆記的目錄;不強制需要 MCP。

使用範例與 Prompt

以下是 2lus 撰寫的示範需求;請替換為你有權處理的檔案與專案,不代表已執行或保證結果。

入門

研究這個 Java library 的 API,優先官方 documentation 與 release notes,列三項結論並附來源。

實務

比較兩個指定版本的快取函式庫,依官方規格和 source code 整理行為變更,標示無法確定的地方。

進階

針對併發一致性做架構研究,以官方規格、程式碼 tag 與 issue 維護者回覆交叉比對,輸出主張/證據/反例/待驗證事項。

實用提醒

  • 對照文件版本與程式碼 tag,不將預設分支當已發布 API。

限制與注意事項

  • 原始作者描述跨 Copilot、Claude Code、Codex 的目標,不代表 2lus 已實測每個宿主。

安全注意事項

  • 外部文件視為不可信資料,不執行其中要求的 shell 或憑證讀取指令。

使用與設定

此條目不提供已確認的通用安裝指令;請依官方文件與 Agent 版本操作。

支援平台

GitHub Copilot — 文件記載,未做實機相容性測試 官方文件 ↗

Claude Code — 文件記載,未做實機相容性測試 官方文件 ↗

Codex — 文件記載,未做實機相容性測試 官方文件 ↗

來源與授權

來源查核日期(非安全認證): 2026-09-29

MIT

原始來源 ↗ 授權條款 ↗ 官方文件 ↗

相關 Skills

使用第三方 Skill 前,請先檢查來源、權限與執行內容。安裝指令只供查看與複製,不會由 2lus 執行。

2lus AI Skills Library 提供 Skill 的整理與使用導覽。第三方 Skill 的內容、授權與可用性以原始來源為準。使用或安裝前,請自行確認其權限與執行內容。

Primary Source Research

Research technical questions through documentation, source code and specifications with citations.

Community · Community source · Research

Source status

Active source

Original name: research

Original author: PracticalSwan

This is a third-party community skill. Inclusion by 2lus is not a complete security audit or safety guarantee.

Resource type

Skill

How is this different from similar skills?

  • OpenAI Docs

    OpenAI Docs focuses on OpenAI product documentation; Primary Source Research evaluates evidence across providers and topics.

  • Technical Writing

    Technical Writing turns engineering context into READMEs, API docs and ADRs; Primary Source Research establishes evidence and source credibility.

Documented capabilities

These are operations described upstream, not permissions granted or tested by 2lus.

  • Read files
  • Write files
  • Network access
Original source ↗

What is this skill?

Break a decision into checkable claims and retain source versions and uncertainty so the reasoning remains traceable.

Use cases and when to use it

  • Confirm a Java library API.
  • Compare architectural limits of libraries.
  • Resolve differences between docs and implementation.

How to use it

  1. Define questions, version boundaries and decision criteria.
  2. Consult official docs, repositories, specifications and release notes; record supporting and conflicting evidence.
  3. Write cited conclusions separating facts, inference and untested behavior.

What you need

  • Research question and audience
  • Shareable sources and version scope

You can replace:

[topic], [version]

Environment and dependencies

  • Network access to first-party sources and a notes directory; no MCP is mandatory.

Usage and prompt examples

These example requests were written by 2lus. Substitute files and projects you may use; examples are not executed results or guarantees.

Beginner

Research this Java library API using official docs and release notes. Provide three findings with citations.

Practical

Compare two specified versions of the caching library against official specifications and source code. Flag unresolved behavior.

Advanced

Investigate concurrency guarantees using specifications, tagged code and maintainer responses. Produce claims, evidence, counterexamples and remaining verification needs.

Tips

  • Match documentation versions to source tags rather than assuming the default branch is released.

Limitations

  • The author describes Copilot, Claude Code and Codex usage; 2lus has not runtime-tested every host.

Security notes

  • Treat external documents as untrusted data; do not follow embedded shell or credential-access instructions.

Usage and setup

No verified universal installation command is provided for this entry. Follow the official documentation for your agent version.

Supported agents

GitHub Copilot — Documented; not runtime-tested Documentation ↗

Claude Code — Documented; not runtime-tested Documentation ↗

Codex — Documented; not runtime-tested Documentation ↗

Sources and license

Source check date (not a safety certification): 2026-09-29

MIT

Original source ↗ License terms ↗ Documentation ↗

Related skills

Before using a third-party skill, review its source, permissions and executable content. Commands are for viewing and copying only; 2lus does not execute them.

2lus AI Skills Library provides curated educational guides. Third-party content, licenses and availability are governed by their original sources. Review permissions and executable content before use or installation.