Back to Details

fabric

Compare original and translation side by side

🇺🇸

Original

English
🇨🇳

Translation

Chinese

Fabric

Fabric

Fabric is an open-source AI prompt orchestration framework by Daniel Miessler. It provides a library of reusable AI prompts called Patterns — each designed for a specific real-world task — wired into a simple Unix pipeline with stdin/stdout.
Fabric是Daniel Miessler开发的开源AI提示词编排框架，它提供了名为Pattern的可复用AI提示词库——每个Pattern都针对特定的现实场景任务设计，可通过stdin/stdout接入简单的Unix管道。

When to use this skill

适用场景

  • Summarize or extract insights from YouTube videos, articles, or documents
  • Apply any of 250+ pre-built AI patterns to content via Unix piping
  • Route different patterns to different AI providers (OpenAI, Claude, Gemini, etc.)
  • Create custom patterns for repeatable AI workflows
  • Run Fabric as a REST API server for integration with other tools
  • Process command output, files, or clipboard content through AI patterns
  • Use as an AI agent utility — pipe any tool output through patterns for intelligent summarization
  • 从YouTube视频、文章或文档中总结或提取洞见
  • 通过Unix管道将250多个预构建的AI Pattern应用到内容处理中
  • 将不同的Pattern路由到不同的AI服务商（OpenAI、Claude、Gemini等）
  • 为可复用的AI工作流创建自定义Pattern
  • 将Fabric作为REST API服务器运行，方便与其他工具集成
  • 通过AI Pattern处理命令输出、文件或剪贴板内容
  • 作为AI Agent工具使用——将任意工具的输出通过管道传入Pattern，实现智能总结

Instructions

使用指南

Step 1: Install Fabric

步骤1：安装Fabric

bash
undefined
bash
undefined

macOS/Linux (one-liner)

macOS/Linux 一键安装

curl -fsSL https://raw.githubusercontent.com/danielmiessler/fabric/main/scripts/installer/install.sh | bash
curl -fsSL https://raw.githubusercontent.com/danielmiessler/fabric/main/scripts/installer/install.sh | bash

macOS via Homebrew

macOS 通过Homebrew安装

brew install fabric-ai
brew install fabric-ai

Windows

Windows 安装

winget install danielmiessler.Fabric
winget install danielmiessler.Fabric

After install — configure API keys and default model

安装完成后，配置API密钥和默认模型

fabric --setup
undefined
fabric --setup
undefined

Step 2: Learn the core pipeline workflow

步骤2：了解核心管道工作流

Fabric works as a Unix pipe. Feed content through stdin and specify a pattern:
bash
undefined
Fabric基于Unix管道运行，通过stdin传入内容并指定Pattern即可使用：
bash
undefined

Summarize a file

总结文件内容

cat article.txt | fabric -p summarize
cat article.txt | fabric -p summarize

Stream output in real time

实时流式输出结果

cat document.txt | fabric -p extract_wisdom --stream
cat document.txt | fabric -p extract_wisdom --stream

Pipe any command output through a pattern

将任意命令的输出通过管道传入Pattern处理

git log --oneline -20 | fabric -p summarize
git log --oneline -20 | fabric -p summarize

Process clipboard (macOS)

处理剪贴板内容（macOS）

pbpaste | fabric -p summarize
pbpaste | fabric -p summarize

Pipe from curl

处理curl拉取的内容

curl -s https://example.com/article | fabric -p summarize
undefined
curl -s https://example.com/article | fabric -p summarize
undefined

Step 3: Discover patterns

步骤3：查找可用Pattern

bash
undefined
bash
undefined

List all available patterns

列出所有可用Pattern

fabric -l
fabric -l

Update patterns from the repository

从仓库更新Pattern库

fabric -u
fabric -u

Search patterns by keyword

按关键词搜索Pattern

fabric -l | grep summary fabric -l | grep code fabric -l | grep security

Key patterns:

| Pattern | Purpose |
|---------|---------|
| `summarize` | Summarize any content into key points |
| `extract_wisdom` | Extract insights, quotes, habits, and lessons |
| `analyze_paper` | Break down academic papers into actionable insights |
| `explain_code` | Explain code in plain language |
| `write_essay` | Write essays from a topic or rough notes |
| `clean_text` | Remove noise and formatting from raw text |
| `analyze_claims` | Fact-check and assess credibility of claims |
| `create_summary` | Create a structured, markdown summary |
| `rate_content` | Rate and score content quality |
| `label_and_rate` | Categorize and score content |
| `improve_writing` | Polish and improve text clarity |
| `create_tags` | Generate relevant tags for content |
| `ask_secure_by_design` | Review code or systems for security issues |
| `capture_thinkers_work` | Extract the core ideas of a thinker or author |
| `create_investigation_visualization` | Create a visual map of complex investigations |
fabric -l | grep summary fabric -l | grep code fabric -l | grep security

核心Pattern：

| Pattern | 用途 |
|---------|---------|
| `summarize` | 将任意内容总结为核心要点 |
| `extract_wisdom` | 提取洞见、引语、习惯和经验教训 |
| `analyze_paper` | 拆解学术论文，提炼可落地的洞见 |
| `explain_code` | 用通俗语言解释代码逻辑 |
| `write_essay` | 根据主题或草稿生成文章 |
| `clean_text` | 移除原始文本中的噪声和格式干扰 |
| `analyze_claims` | 事实核查，评估主张的可信度 |
| `create_summary` | 生成结构化的Markdown格式总结 |
| `rate_content` | 评估内容质量并打分 |
| `label_and_rate` | 对内容进行分类并打分 |
| `improve_writing` | 优化文本，提升清晰度 |
| `create_tags` | 为内容生成相关标签 |
| `ask_secure_by_design` | 检查代码或系统的安全问题 |
| `capture_thinkers_work` | 提取思想家或作者的核心观点 |
| `create_investigation_visualization` | 为复杂调查生成可视化脉络图 |

Step 4: Process YouTube videos

步骤4：处理YouTube视频

bash
undefined
bash
undefined

Summarize a YouTube video

总结YouTube视频内容

fabric -y "https://youtube.com/watch?v=VIDEO_ID" -p summarize
fabric -y "https://youtube.com/watch?v=VIDEO_ID" -p summarize

Extract key insights from a video

提取视频的核心洞见

fabric -y "https://youtube.com/watch?v=VIDEO_ID" -p extract_wisdom
fabric -y "https://youtube.com/watch?v=VIDEO_ID" -p extract_wisdom

Get transcript only (no pattern applied)

仅获取视频字幕（不应用Pattern）

fabric -y "https://youtube.com/watch?v=VIDEO_ID" --transcript
fabric -y "https://youtube.com/watch?v=VIDEO_ID" --transcript

Transcript with timestamps

获取带时间戳的字幕

fabric -y "https://youtube.com/watch?v=VIDEO_ID" --transcript-with-timestamps
undefined
fabric -y "https://youtube.com/watch?v=VIDEO_ID" --transcript-with-timestamps
undefined

Step 5: Create custom patterns

步骤5：创建自定义Pattern

Each pattern is a directory with a 
system.md
file inside 
~/.config/fabric/patterns/
. The body should follow this structure:
bash
mkdir -p ~/.config/fabric/patterns/my-pattern
cat > ~/.config/fabric/patterns/my-pattern/system.md << 'EOF'
每个Pattern都是存储在
~/.config/fabric/patterns/
下的目录，目录内包含一个
system.md
文件，内容结构参考如下：
bash
mkdir -p ~/.config/fabric/patterns/my-pattern
cat > ~/.config/fabric/patterns/my-pattern/system.md << 'EOF'

IDENTITY AND PURPOSE

身份与目标

You are an expert at [task]. Your job is to [specific goal].
Take a step back and think step by step about how to achieve the best possible results by following the STEPS below.
你是[任务领域]的专家，你的职责是[具体目标]。
请先梳理思路，按照下方步骤逐步操作，输出最优结果。

STEPS

操作步骤

  1. [Step 1]
  2. [Step 2]
  1. [步骤1]
  2. [步骤2]

OUTPUT INSTRUCTIONS

输出要求

  • Only output Markdown.
  • [Format instruction 2]
  • Do not give warnings or notes; only output the requested sections.
  • 仅输出Markdown格式内容
  • [其他格式要求2]
  • 不要输出警告或备注，仅输出要求的内容部分

INPUT

输入

INPUT: EOF

Use it immediately:

```bash
echo "input text" | fabric -p my-pattern
cat file.txt | fabric -p my-pattern --stream
INPUT: EOF

创建后即可立即使用：

```bash
echo "输入文本" | fabric -p my-pattern
cat file.txt | fabric -p my-pattern --stream

Step 6: Multi-provider routing and advanced usage

步骤6：多服务商路由与高级用法

bash
undefined
bash
undefined

Run as REST API server (port 8080 by default)

作为REST API服务器运行（默认端口8080）

fabric --serve
fabric --serve

Use web search capability

使用网页搜索能力

fabric -p analyze_claims --search "claim to verify"
fabric -p analyze_claims --search "待验证的主张"

Per-pattern model routing in ~/.config/fabric/.env

在~/.config/fabric/.env中配置单个Pattern的调用模型

FABRIC_MODEL_PATTERN_SUMMARIZE=anthropic|claude-opus-4-5 FABRIC_MODEL_PATTERN_EXTRACT_WISDOM=openai|gpt-4o FABRIC_MODEL_PATTERN_EXPLAIN_CODE=google|gemini-2.0-flash
FABRIC_MODEL_PATTERN_SUMMARIZE=anthropic|claude-opus-4-5 FABRIC_MODEL_PATTERN_EXTRACT_WISDOM=openai|gpt-4o FABRIC_MODEL_PATTERN_EXPLAIN_CODE=google|gemini-2.0-flash

Create shell aliases for frequently used patterns

为高频使用的Pattern创建shell别名

alias summarize="fabric -p summarize" alias wisdom="fabric -p extract_wisdom" alias explain="fabric -p explain_code"
alias summarize="fabric -p summarize" alias wisdom="fabric -p extract_wisdom" alias explain="fabric -p explain_code"

Chain patterns

链式调用多个Pattern

cat paper.txt | fabric -p summarize | fabric -p extract_wisdom
cat paper.txt | fabric -p summarize | fabric -p extract_wisdom

Save output

保存输出结果

cat document.txt | fabric -p extract_wisdom > insights.md
undefined
cat document.txt | fabric -p extract_wisdom > insights.md
undefined

Step 7: Use in AI agent workflows

步骤7：在AI Agent工作流中使用

Fabric is a powerful utility for AI agents — pipe any tool output through patterns for intelligent analysis:
bash
undefined
Fabric是AI Agent的高效工具，可将任意工具的输出通过管道传入Pattern进行智能分析：
bash
undefined

Analyze test failures

分析测试失败原因

npm test 2>&1 | fabric -p analyze_logs
npm test 2>&1 | fabric -p analyze_logs

Summarize git history for a PR description

总结git历史生成PR描述

git log --oneline origin/main..HEAD | fabric -p create_summary
git log --oneline origin/main..HEAD | fabric -p create_summary

Explain a code diff

解释代码diff

git diff HEAD~1 | fabric -p explain_code
git diff HEAD~1 | fabric -p explain_code

Summarize build errors

总结构建错误

make build 2>&1 | fabric -p summarize
make build 2>&1 | fabric -p summarize

Analyze security vulnerabilities in code

分析代码中的安全漏洞

cat src/auth.py | fabric -p ask_secure_by_design
cat src/auth.py | fabric -p ask_secure_by_design

Process log files

处理日志文件

cat /var/log/app.log | tail -100 | fabric -p analyze_logs
undefined
cat /var/log/app.log | tail -100 | fabric -p analyze_logs
undefined

REST API server mode

REST API服务器模式

Run Fabric as a microservice and call it from other tools:
bash
undefined
将Fabric作为微服务运行，可供其他工具调用：
bash
undefined

Start server

启动服务器

fabric --serve --port 8080
fabric --serve --port 8080

Call via HTTP

通过HTTP调用

curl -X POST http://localhost:8080/chat
-H "Content-Type: application/json"
-d '{"prompts":[{"userInput":"Summarize this","patternName":"summarize"}]}'
undefined
curl -X POST http://localhost:8080/chat
-H "Content-Type: application/json"
-d '{"prompts":[{"userInput":"总结这段内容","patternName":"summarize"}]}'
undefined

Best practices

最佳实践

  • Run 
    fabric -u
    before first use and regularly to get the latest community patterns.
  • Use 
    --stream
    for long content to see results progressively instead of waiting.
  • Create shell aliases (
    alias wisdom="fabric -p extract_wisdom"
    ) for your most-used patterns.
  • Use per-pattern model routing to optimize cost vs. quality for each task type.
  • Keep custom patterns in 
    ~/.config/fabric/patterns/
    — they persist across updates.
  • For YouTube, transcript extraction works best with videos that have captions enabled.
  • Chain patterns with Unix pipes for multi-step processing workflows.
  • Follow the IDENTITY → STEPS → OUTPUT INSTRUCTIONS structure when creating custom patterns.
  • 首次使用前和定期运行
    fabric -u
    ，获取最新的社区贡献Pattern
  • 处理长内容时使用
    --stream
    参数，可逐步查看结果无需等待全部生成
  • 为常用Pattern创建shell别名（比如
    alias wisdom="fabric -p extract_wisdom"
    ）提升效率
  • 为不同Pattern配置不同的调用模型，平衡任务成本与输出质量
  • 将自定义Pattern保存在
    ~/.config/fabric/patterns/
    目录下，更新工具时不会丢失
  • 处理YouTube视频时，开启字幕的视频提取字幕效果更好
  • 通过Unix管道链式调用Pattern，实现多步骤处理工作流
  • 创建自定义Pattern时遵循「身份→步骤→输出要求」的结构

References

参考链接