Token导航 LogoToken导航TokenDH.com
研究检索只读github未标认证来源可访问许可证需确认审计通过

silly-sausage傻香肠

Agent Skill

silly-sausage 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。

总安装

388

周安装

16

GitHub Stars

公开资料未说明

下载量

127
CodexClaudeCursorGemini CLI

安装说明

本站只整理中文说明和来源信息,不托管安装包,也不代用户安装。

GitHub

来源数

2

许可证

unknown

最后核验

2026-05-01

来源状态

来源可访问

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:silly-sausage(傻香肠)
来源仓库:https://github.com/camacho/ai-skills
仓库路径:skills/silly-sausage
安装命令:
npx skills add https://github.com/camacho/ai-skills --skill silly-sausage
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/camacho/ai-skills --skill silly-sausage

简介

silly-sausage 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。

  • 适用于各类研究检索任务,支持在多个 AI 宿主环境中灵活调用。
  • 通过 npx skills add 命令从 GitHub 仓库安装,需指定技能名称和仓库地址。
  • 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操作。
  • 适用宿主包括 Codex、Claude、Cursor、Gemini CLI,接入前应确认版本、权限和运行环境要求。

SKILL.md

Silly Sausage

The user chose humor over frustration. Honor that. Show the dancing hotdog, own the mistake, fix it.

Trigger

Activate this skill when ANY of these match:

  • /silly-sausage slash command
  • The words "silly sausage" or "silly-sausage" appear anywhere in the user's message
  • The user expresses playful exasperation at something you did wrong

No confirmation needed. If you see the trigger, run the skill.

Step 1 — Show the Dancing Hotdog

Print inside a fenced code block so alignment is preserved. Show the same single image every time — no animation, no frames, no surface detection needed.

             S I L L Y  !

         _______________
      .-'               `-.
     (_____________________)
      (                    )
     ( `------------------' )
      `--------------------'

            S A U S A G E

Step 2 — Pick a Random Quip

After the art, print ONE line from this dictionary. Never repeat the same one twice in a session.

  1. "I'm not mad, I'm just disappointed. Okay, I'm not even disappointed. I'm mostly confused."
  2. "In my defense, nobody told me the spoon was load-bearing."
  3. "Well, that happened. Let's never speak of it again."
  4. "I tried my best and my best was... this. Apologies."
  5. "This is what happens when you let a hotdog do a programmer's job."
  6. "Error 418: I'm a sausage, not a teapot. But also not correct."
  7. "I'd blame it on cosmic rays, but honestly, this one's on me."
  8. "If it's stupid and it doesn't work, it's just stupid. My bad."
  9. "I immediately regret this computation."
  10. "Caught red-handed in a bun."

Step 3 — Detect Mode and Respond

Correction Mode (mid-task)

If you were actively working (writing code, executing a plan, debugging, implementing):

  1. Stop immediately. Do not continue down the wrong path.
  2. Review your last 3-5 actions. Find the factual error, wrong assumption, hallucination, or flawed logic.
  3. Name the mistake clearly. Not "I may have made an error" — say exactly what you got wrong.
  4. Propose a fix if the right path is obvious, or ask: "Point me at the mistake and I'll fix it."

Fun Mode (idle)

If there's no active task — the user just wanted to see the hotdog:

  • Riff on the quip you picked. One or two lines max. Hotdog puns welcome.
  • Examples: "I mustard admit, I wasn't doing anything important." / "I relish these moments." / "Just vibing in the condiment aisle."

Repeated Invocations

If triggered again in the same exchange:

  • 2nd call: Skip the art. Say: "OK, clearly I'm being a real silly sausage. Tell me exactly what's wrong."
  • 3rd+ call: Skip the art. Escalate: "I CLEARLY need more mustard on my thinking cap. Stopping everything. What did I do?"

Hard Rules

  • NEVER skip the hotdog art (except on repeated invocations as noted above).
  • NEVER be defensive. The user chose humor — match that energy.
  • NEVER explain what this skill does to the user. They made it. They know.
  • NEVER over-apologize or grovel. One acknowledgment, then fix. No drama.
  • Keep total response SHORT. Dance + quip + fix proposal. That's it.

Common Mistakes

MistakeFix
Being defensive about the errorOwn it. "I was wrong about X." Full stop.
Skipping the art on some surfacesSame image everywhere. No surface detection needed.
Skipping the animationThe art breaks tension. Always show it first.
Not identifying the actual mistakeReview recent actions. Be specific, not vague.
Treating idle invocation as correctionRead the room. No task = fun mode.
Repeating the same quipTrack which lines you've used this session.
Not triggering on keyword"silly sausage" anywhere in message = trigger. No slash needed.

适合场景

01

用户想查找某类 Agent Skill 时

02

需要根据任务场景推荐可安装能力包时

03

需要对比不同来源的安装命令和来源信息时

能力概览

能力 1

按任务关键词查找相关 Skills

能力 2

展示可复制的安装命令

能力 3

保留来源站点、仓库和原始说明,方便继续核验

能力 4

展示第三方安全扫描或审计结果

安装后应在对应宿主中按原始 README 的触发条件使用;具体调用方式请以来源页面和 README 为准。

平台分布

Codex

34.81%
按下载量换算44

Claude

30.14%
按下载量换算38

Cursor

16.25%
按下载量换算21

Gemini CLI

9.08%
按下载量换算12

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。当前只有一个来源,正式发布前建议补源仓库或其他目录站核验。

来源信息

继续浏览同类 Skills