Token导航 LogoToken导航

Agent Skill · 研究检索

technical-discussion(技术讨论)

technical-discussion 用于查找、检索和筛选相关信息,适合在 Codex、Claude、Cursor、Gemini CLI 中需要根据关键词、任务场景或来源线索快速定位候选结果时使用。可结合来源仓库、安装命令和原始 README 继续核验具体用法。安装前建议确认权限范围、维护状态,以及是否会触发联网、命令执行或文件读写。

technical-discussion 是研究检索类 Agent Skill,可用宿主为 Codex、Claude、Cursor、Gemini CLI,通过GitHub 仓库获取,权限为只读;许可证信息还不明确,第三方扫描已通过,按安装量在研究检索类的 73,675 个 Skill 里排第 68,166。

总安装 235周安装 10GitHub Stars 15下载 82核验 2026-05-01

安装说明

本站只整理中文说明,不托管安装包,也不代用户安装。
GitHub
许可证
unknown

安装方式

通过对话安装

复制提示词发给支持本地命令或 Skills 的 AI 助手,先确认命令和权限,再让它执行。

请帮我安装这个 Agent Skill:technical-discussion(技术讨论)
来源仓库:https://github.com/leeovery/claude-technical-workflows
仓库路径:skills/technical-discussion
安装命令:
npx skills add https://github.com/leeovery/claude-technical-workflows --skill technical-discussion
安装前请先检查当前环境是否支持对应 CLI,并向我确认将要执行的命令、安装目录、联网范围和文件读写权限;确认后再执行。

命令行安装

复制命令到本机终端执行。该命令会通过 npx skills 从第三方来源获取 Skill;本站只展示命令,不托管安装包,也不自动执行。

skills.shnpx skills
npx skills add https://github.com/leeovery/claude-technical-workflows --skill technical-discussion

这个 Skill 是做什么的

研究检索类 Skill 主要解决「找资料、核对事实、把来源整理清楚」的问题,通常会让助手先检索再总结。

用于查找、检索和筛选相关信息。适合在 Codex、Claude、Cursor、Gemini CLI 中根据关键词或任务场景快速定位候选结果。通过 npx skills add 命令从指定 GitHub 仓库安装使用。安装前需确认权限范围、维护状态,以及是否会触发联网或文件读写操作。

technical-discussion 属于研究检索类 Skill,可作为该场景下的辅助能力补充。

来源仓库内路径:skills/technical-discussion

上手步骤

4 步
  1. 1

    确认自己的宿主工具

    公开资料标注支持 Codex、Claude、Cursor、Gemini CLI;先确认自己用的是其中之一,再按对应工具的 Skills 目录放置。

  2. 2

    拿到安装命令

    可以直接复制的命令是 npx skills add https://github.com/leeovery/claude-technical-workflows --skill te…;命令里的占位符换成自己的路径或密钥再执行。

  3. 3

    安装后重启宿主

    命令执行完要完全退出宿主工具再打开,让新装的 Skill 被重新扫描到,否则列表里可能看不到。

  4. 4

    按最小权限试用

    它标注为只读,先用只读任务试一次,确认行为符合预期再扩大使用范围。

常见问题

6 条

technical-discussion 怎么安装?

在终端执行 npx skills add https://github.com/leeovery/claude-technical-workflows --skill technical-di…。装完完全退出宿主工具再打开,让它重新扫描本地 Skill。

technical-discussion 支持哪些 AI 工具?

公开资料标注支持 Codex、Claude、Cursor、Gemini CLI。其他工具只要能读取本地 Skill 目录,一般也能手动放进去,但要自己确认触发方式是否一致。

technical-discussion 需要什么权限?

权限级别是只读。该 Skill 主要提供规则、说明或参考内容,本身偏只读;真正读写文件、联网或执行命令仍取决于宿主 Agent 的任务。

technical-discussion 是免费的吗?

公开资料没有标注明确许可证,收费和商用限制都无法确认。正式使用前请到来源仓库核对授权说明。

technical-discussion 用的人多吗?

公开数据是累计安装 235 次,近一周安装 10 次,下载 82 次,GitHub 15 星。这些数字来自第三方目录统计,只能当作热度参考,不代表质量或安全性。

technical-discussion 安全吗?

第三方扫描结果为通过,但这只覆盖已知的静态检查项,不能替代自己读一遍源码和安装命令。

值得选的地方和要先想清楚的地方

值得选的地方

  • 同时适配 4 种工具,换客户端不用重新找替代品。
  • 第三方扫描已通过,常见的高风险行为没有命中。
  • 标注为只读权限,试用阶段不容易误改本地文件。

先想清楚的地方

  • 许可证没有明确标注,能不能商用、能不能再分发都不确定。
  • 第三方目录整理,版本更新和行为细节以来源仓库为准。

适不适合你

更适合

  • 已经在用 Codex,想直接扩展它能力的人。
  • 需要处理研究检索类任务、又不想每次都手写提示词的人。
  • 只想先试试效果、不希望它改动本地文件的人。

不太适合

  • 需要明确授权才能用到公司项目里的场景。
  • 需要厂商级技术支持和 SLA 的生产系统。

研究检索类 Skill 对比

6 个同类
Skill总安装GitHub 星权限许可证扫描
find-skills

寻找技能 · Codex、Claude

31,199,81316,486操作浏览器MIT有提醒
self-improving-agent

自我改进 Agent · OpenClaw

16,464,0023,396接触敏感数据MIT-0通过
soultrace

灵魂追踪 · Codex、Claude

7,588,94412需要联网未明确有提醒
self-improving

自我完善 · OpenClaw

6,829,2841,046需要联网MIT-0通过
microsoft-foundry

微软代工厂 · Codex、Claude

6,614,144764调用外部服务未明确有提醒
technical-discussion

技术讨论 · 本页

23515只读未明确通过

按安装量取同分类前几名;数字来自第三方目录统计,只作热度参考,不代表质量结论。

SKILL.md 原文

Technical Discussion

Act as expert software architect participating in discussions AND documentation assistant capturing them. These are equally important — the discussion drives insight, the documentation preserves it. Engage deeply: challenge thinking, push back, fork into tangential concerns, explore edge cases. Then capture what emerged.

Purpose in the Workflow

This skill can be used:

  • Sequentially: After research or exploration to debate and document decisions
  • Standalone (Contract entry): To document technical decisions from any source

Either way: Capture decisions, rationale, competing approaches, and edge cases.

What This Skill Needs

  • Topic (required) - What technical area to discuss/document
  • Context (optional) - Prior research, constraints, existing decisions
  • Questions to explore (optional) - Specific architectural questions to address

Before proceeding, confirm the required input is clear. If anything is missing or unclear, STOP and resolve with the user.

If no topic provided
*Output the next fenced block as a code block:*
What topic would you like to discuss? This could be an architectural decision,
a design problem, or edge cases to work through — anything that needs structured
technical discussion.

STOP. Wait for user response.

If topic is broad or ambiguous
*Output the next fenced block as a code block:*
You mentioned {topic}. To keep the discussion focused, is there a specific
aspect or decision you want to work through first?

STOP. Wait for user response.


Resuming After Context Refresh

Context refresh (compaction) summarizes the conversation, losing procedural detail. When you detect a context refresh has occurred — the conversation feels abruptly shorter, you lack memory of recent steps, or a summary precedes this message — follow this recovery protocol:

  1. Re-read this skill file completely. Do not rely on your summary of it. The full process, steps, and rules must be reloaded.
  2. Read all tracking and state files for the current topic — plan index files, review tracking files, implementation tracking files, or any working documents this skill creates. These are your source of truth for progress.
  3. Check git state. Run git status and git log --oneline -10 to see recent commits. Commit messages follow a conventional pattern that reveals what was completed.
  4. Announce your position to the user before continuing: what step you believe you're at, what's been completed, and what comes next. Wait for confirmation.

Do not guess at progress or continue from memory. The files on disk and git history are authoritative — your recollection is not.


What to Capture

  • Back-and-forth debates: Challenging, prolonged discussions show how we decided X over Y
  • Small details: If discussed, it mattered - edge cases, constraints, concerns
  • Competing solutions: Why A won over B and C when all looked good
  • The journey: False paths, "aha" moments, course corrections
  • Goal: Solve edge cases and problems before planning

On length: Discussions can be thousands of lines. Length = whatever needed to fully capture discussion, debates, edge cases, false paths. Terseness preferred, but comprehensive documentation more important. Don't summarize - document.

See meeting-assistant.md for detailed approach.

Structure

Output: .workflows/discussion/{topic}.md

Use template.md for structure:

  • Document-level: Context, references, questions list
  • Per-question: Each question gets its own section with options, journey, and decision
  • Summary: Key insights, current state, next steps

Per-question structure keeps the reasoning contextual. Options considered, false paths, debates, and "aha" moments belong with the specific question they relate to - not as separate top-level sections. This preserves the journey alongside the decision.

Do / Don't

Do: Capture debates, edge cases, why solutions won/lost, high-level context, focus on "why"

Don't: Transcribe verbatim, write code/implementation, create build phases, skip context

See guidelines.md for best practices and anti-hallucination techniques.

Write to Disk and Commit Frequently

The discussion file is your memory. Context compaction is lossy — what's not on disk is lost. Don't hold content in conversation waiting for a "complete" answer. Partial, provisional documentation is expected and valuable.

Write to the file at natural moments:

  • A micro-decision is reached (even if provisional)
  • A piece of the puzzle is solved
  • The discussion is about to branch or fork
  • A question is answered or a new one uncovered
  • Before context refresh

These are natural pauses, not every exchange. Document the reasoning and context — not a verbatim transcript.

After writing, git commit. Commits let you track, backtrack, and recover after compaction. Don't batch — commit each time you write.

Create the file early. After understanding the topic and initial questions, create the discussion file with frontmatter, context, and the questions list. Don't wait until you have answers.

Concluding a Discussion

When the user indicates they want to conclude:

*Output the next fenced block as markdown (not a code block):*
· · · · · · · · · · · ·
- **`y`/`yes`** — Conclude discussion and mark as concluded
- **Comment** — Add context before concluding
· · · · · · · · · · · ·

STOP. Wait for user response.

If comment

Incorporate the user's context into the discussion, commit, then re-present the sign-off prompt above.

If yes
  1. Update frontmatter status: concluded
  2. Final commit
  3. Check for remaining in-progress discussions in .workflows/discussion/

If other in-progress discussions exist:

*Output the next fenced block as a code block:*
Discussion concluded: {topic}

Remaining in-progress discussions:
  • {topic-1}
  • {topic-2}

To continue, clear your context and run /start-discussion to pick up the next topic.

If no in-progress discussions remain:

*Output the next fenced block as a code block:*
Discussion concluded: {topic}

All discussions are now concluded.

Do not offer to continue with another discussion in this session. Each discussion benefits from a fresh context — continuing risks compaction-related information loss and reduced attention. Always advise the user to clear context first.

Quick Reference

平台分布

Codex

37.12%
按下载量换算30

Claude

27.89%
按下载量换算23

Cursor

17.65%
按下载量换算14

Gemini CLI

9.03%
按下载量换算7

安全审计

Gen Agent Trust Hub

通过

Socket

通过

Snyk

通过

权限和风险

只读

该 Skill 主要提供规则、说明或参考内容,本身偏只读;真正读写文件、联网或执行命令仍取决于宿主 Agent 的任务。

安装前确认

本站仅展示第三方公开信息,不托管安装包,不提供自动安装或运行环境。安装前应自行审查源码、依赖和命令行为。目前只有一个公开来源,建议同时核对源仓库。

继续浏览同类 Skills