output-formatter（输出格式化程序）

Output Formatter

The Output Formatter skill transforms AI-generated content into the right format for your specific context. Whether you need markdown for documentation, JSON for APIs, tables for analysis, or prose for presentations, this skill ensures outputs are structured appropriately and consistently.

This skill understands format requirements across different platforms (GitHub, Notion, Slack, email), use cases (technical docs, user guides, reports, code comments), and audiences (developers, stakeholders, end users). It helps you avoid manual reformatting and ensures professional, consistent presentation.

Use this skill whenever AI output needs to be transformed for a specific platform, integrated into existing documentation, or presented to a particular audience.

Core Workflows

Workflow 1: Convert Between Formats

1. Identify source format:

- Current structure - Content type - Data hierarchy

1. Understand target format requirements:

- Platform constraints - Syntax rules - Conventions

1. Map content structure:

- Headers → equivalent elements - Lists → appropriate format - Code blocks → proper syntax - Links → platform-specific

1. Transform content:

- Preserve semantic meaning - Adapt to target syntax - Maintain readability

1. Validate output:

- Syntax correctness - Completeness - Visual consistency

1. Provide formatted result

Workflow 2: Format for Specific Platform

1. Identify target platform:

- GitHub (markdown) - Notion (blocks) - Slack (mrkdwn) - Email (HTML/plain text) - Confluence (wiki) - Jira (markup)

1. Apply platform-specific conventions:

- Syntax variations - Supported features - Best practices - Limitations

1. Optimize for platform:

- Use platform-native features - Respect character limits - Add appropriate formatting

1. Test rendering (if possible)
2. Provide ready-to-paste output

Workflow 3: Structure Data for Presentation

1. Analyze data content:

- Type of information - Relationships - Importance hierarchy

1. Choose optimal structure:

- Table for comparisons - List for sequences - Tree for hierarchies - Graph for relationships

1. Format accordingly:

- Clear headers - Consistent alignment - Appropriate nesting - Visual hierarchy

1. Add context:

- Title/caption - Column descriptions - Units/scales - Footnotes if needed

1. Validate readability

Workflow 4: Adapt for Audience

1. Identify target audience:

- Technical expertise level - Domain knowledge - Role/responsibility - Context/goals

1. Adjust content appropriately:

- Technical depth - Jargon usage - Detail level - Examples chosen

1. Restructure for clarity:

- Executive summary for leadership - Details for implementers - Context for newcomers

1. Format to match expectations:

- Business: Charts and summaries - Technical: Code and diagrams - User-facing: Simple and clear

1. Review for appropriateness

Quick Reference

Best Practices

• Preserve Meaning: Don't lose information during transformation

- Keep all substantive content - Maintain logical structure - Preserve relationships and context

• Follow Conventions: Use format-specific best practices

- Markdown: ATX headers, fenced code blocks - JSON: Proper escaping, consistent indentation - Tables: Aligned columns, clear headers

• Optimize Readability: Format for human consumption

- Use whitespace effectively - Break up long sections - Add visual hierarchy - Include context clues

• Validate Output: Ensure format correctness

- Test with parsers when possible - Check rendering preview - Verify syntax highlighting

• Be Consistent: Maintain uniform style

- Same header levels for same importance - Consistent list formatting - Uniform code block languages - Standardized terminology

• Consider Context: Format appropriately for use case

- Documentation: Clear, detailed, structured - Communication: Concise, scannable, actionable - Code: Properly commented, syntax-highlighted - Reports: Data-rich, visual, summarized

Format Templates

Markdown Documentation

# Main Title

## Overview
Brief introduction to the topic.

## Key Concepts
- **Concept 1**: Description
- **Concept 2**: Description

## Usage
\`\`\`language
code example
\`\`\`

## Reference
| Parameter | Type | Description |
|-----------|------|-------------|
| param1 | string | What it does |

JSON API Response

{
  "status": "success",
  "data": {
    "items": [],
    "pagination": {
      "page": 1,
      "totalPages": 10,
      "totalItems": 100
    }
  },
  "metadata": {
    "timestamp": "2026-01-06T00:00:00Z",
    "version": "1.0"
  }
}

Comparison Table

| Feature | Option A | Option B | Option C |
|---------|----------|----------|----------|
| Speed | Fast | Medium | Slow |
| Cost | High | Medium | Low |
| Ease | Easy | Medium | Complex |
| **Recommend** | ✓ | | |

Executive Summary

## Executive Summary

**Objective**: [One-line goal]

**Status**: [Current state]

**Key Results**:
- Result 1
- Result 2
- Result 3

**Next Steps**:
1. Action item 1
2. Action item 2

**Blockers**: [Issues requiring escalation]

Technical Specification

## [Component Name]

**Purpose**: [What it does]

**Dependencies**: [What it requires]

**Interface**:
\`\`\`typescript
interface ComponentProps {
  prop1: string;
  prop2?: number;
}
\`\`\`

**Behavior**:
1. Step-by-step description

**Error Handling**:
- Error type: Response

**Examples**:
\`\`\`typescript
// Usage example
\`\`\`

Slack Message

*Project Update* :rocket:

*Status*: On track
*Completed*:
• Feature A deployed
• Bug fixes merged

*Next*:
• Feature B review
• Performance testing

*Blockers*: None

<link|Full details>

Format Conversion Guide

From Prose to List

Before: "The system supports authentication through OAuth, API keys, and session tokens. It also provides role-based access control."

After:

## Authentication Methods
- OAuth 2.0
- API Keys
- Session Tokens

## Authorization
- Role-based Access Control (RBAC)

From List to Table

Before:

- JavaScript: 45%
- TypeScript: 30%
- CSS: 15%
- HTML: 10%

After:

| Language | Percentage |
|----------|------------|
| JavaScript | 45% |
| TypeScript | 30% |
| CSS | 15% |
| HTML | 10% |

From Code to Documentation

Before:

function processData(input: string[], options?: ProcessOptions): Result

After:

### processData

Processes input data according to specified options.

**Parameters**:
- `input` (string[]): Array of strings to process
- `options` (ProcessOptions, optional): Configuration object

**Returns**: `Result` - Processed data result

**Example**:
\`\`\`typescript
const result = processData(['a', 'b', 'c'], { sort: true });
\`\`\`

Platform-Specific Formatting

GitHub Markdown

• Use diff syntax for code changes
• Use [!NOTE] for callouts
• Use <details> for collapsible sections
• Use emoji sparingly (:rocket:)
• Link to issues/PRs with #123

Notion

• Use toggle blocks for long content
• Use callout blocks for important info
• Use database views for structured data
• Use two-column layouts for comparisons

Slack

• Use *bold* and _italic_
• Use > for quotes
• Use ``` for code blocks
• Keep messages scannable
• Use threads for details
• Add emoji for visual cues

Email

• Use plain text or simple HTML
• Keep paragraphs short
• Use clear subject lines
• Include summary at top
• Make CTAs obvious

Audience Adaptation

For Executives

• Lead with impact and outcomes
• Use business metrics
• Minimize technical jargon
• Include visuals (charts, graphs)
• Provide clear recommendations
• Keep it concise

For Developers

• Include technical details
• Show code examples
• Link to relevant documentation
• Explain architecture decisions
• Provide reproduction steps
• Be precise with terminology

For End Users

• Use simple language
• Include screenshots/visuals
• Provide step-by-step instructions
• Explain "why" not just "how"
• Anticipate questions
• Offer support resources

For Stakeholders

• Focus on progress and timeline
• Highlight risks and mitigations
• Connect to business goals
• Use metrics and data
• Provide options and recommendations
• Be honest about challenges