AGENT SKILL

api-documentation

Creates comprehensive API documentation including endpoints, parameters, authentication, request/response schemas, and code examples. Use when: documenting REST APIs, GraphQL interfaces, SDKs, or developer-facing technical specifications.

License 许可
AGPL-3.0
最近更新
2026-11-14
Stars
★ 25
Forks
9

01 前言 Front Matter

name
api-documentation
owner
pm
scope
common
status
active
version
1.0.2
metadata
{ "type": "implementation", "triggers": [ "api documentation", "document api", "api reference", "developer documentation", "rest api docs", "graphql docs", "sdk documentation" ] }
description
Creates comprehensive API documentation including endpoints, parameters, authentication, request/response schemas, and code examples. Use when: documenting REST APIs, GraphQL interfaces, SDKs, or developer-facing technical specifications.
gemini-parity
skip
last_reviewed
1794096000000
prerequisites
none

03 SKILL.md 正文

Context

This skill provides systematic API documentation capabilities, creating comprehensive developer resources that enable API integration and adoption. It ensures documentation is accurate, complete, and developer-friendly.

When to Use

API Documentation:

  • Trigger: "Document the API" or "Create API reference"
  • Use Case: Creating comprehensive documentation for REST/GraphQL APIs

SDK Documentation:

  • Trigger: "Document SDK" or "Create library documentation"
  • Use Case: Documenting software development kits and client libraries

API Updates:

  • Trigger: "Update API docs for new endpoint" or "Document breaking changes"
  • Use Case: Keeping documentation current with API changes

Developer Resources:

  • Trigger: "Create developer guide" or "Write integration tutorial"
  • Use Case: Helping developers integrate with APIs successfully

Execution Steps

Step 1: API Analysis and Structure

Purpose: Understand API structure and documentation requirements.

API Analysis Process:

  1. API Type Identification:

    • REST API (resources, HTTP methods, status codes)
    • GraphQL API (schema, queries, mutations)
    • RPC/GraphQL (procedure calls)
    • WebSocket API (real-time communication)
    • SDK/Library (classes, methods, parameters)
  2. Endpoint/Operation Inventory:

    • List all endpoints or operations
    • Group by resource or functionality
    • Identify authentication requirements
    • Note rate limiting or throttling
  3. Data Model Analysis:

    • Request schemas and parameters
    • Response schemas and formats
    • Data types and validation rules
    • Error response formats
  4. Developer Use Cases:

    • Common integration scenarios
    • Typical workflows and sequences
    • Authentication flows
    • Error handling examples

Documentation Structure:

1. Overview (getting started, authentication)
2. Quick Start (simple example)
3. API Reference (detailed endpoints)
4. Guides (tutorials, best practices)
5. SDK Reference (if applicable)
6. Error Handling (troubleshooting)
7. Changelog (version history)

Step 2: Getting Started Documentation

Purpose: Help developers begin using the API quickly.

Getting Started Components:

  1. Overview Section:

    • What the API does
    • Key features and capabilities
    • Use cases and benefits
    • Access requirements (API keys, OAuth)
  2. Authentication Guide:

    • Authentication methods (API key, OAuth, JWT)
    • How to obtain credentials
    • How to include credentials in requests
    • Token refresh procedures
  3. Quick Start Example:

    • Simple, working example
    • Minimal required parameters
    • Expected response
    • Language-specific examples (Python, JavaScript, cURL)
  4. Setup Instructions:

    • Environment setup
    • Dependency installation
    • Configuration steps
    • First request walkthrough

Quick Start Template:

## Quick Start

### 1. Get API Key
Register at [url] to get your API key

### 2. Make Your First Request
\`\`\`bash
curl -X GET "https://api.example.com/v1/users" \\
  -H "Authorization: Bearer YOUR_API_KEY"
\`\`\`

### 3. Expected Response
\`\`\`json
{
  "users": [...],
  "total": 10
}
\`\`\`

Step 3: API Reference Documentation

Purpose: Create detailed documentation for each endpoint/operation.

Endpoint Documentation Structure:

  1. Endpoint Overview:

    • HTTP method and path
    • Description of what it does
    • Use case example
    • Rate limiting info
  2. Request Parameters:

    • Path parameters (required)
    • Query parameters (optional)
    • Request body schema
    • Headers required
    • Authentication requirements
  3. Response Documentation:

    • Success response schema
    • Example responses
    • Status codes and meanings
    • Pagination info (if applicable)
  4. Error Responses:

    • Error response format
    • Common error codes
    • Troubleshooting steps
    • Error examples

Endpoint Template:

## Get Users

Retrieves a list of users with optional filtering.

### Request
\`GET /v1/users\`

### Query Parameters
| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| limit | integer | No | Number of results (1-100, default: 20) |
| offset | integer | No | Pagination offset (default: 0) |

### Response
\`\`\`json
{
  "users": [
    {
      "id": "user_123",
      "name": "John Doe",
      "email": "john@example.com"
    }
  ],
  "total": 1,
  "limit": 20,
  "offset": 0
}
\`\`\`

### Errors
| Status | Description |
|--------|-------------|
| 400 | Invalid query parameter |
| 401 | Unauthorized (invalid API key) |
| 429 | Rate limit exceeded |

Step 4: Request/Response Schema Documentation

Purpose: Document data structures and validation rules.

Schema Documentation:

  1. Request Schemas:

    • JSON Schema or TypeScript interfaces
    • Required vs. optional fields
    • Validation rules
    • Constraints and ranges
    • Example requests
  2. Response Schemas:

    • Response object structure
    • Field descriptions and types
    • Nullable fields
    • Enum values
    • Example responses
  3. Data Types:

    • Primitives (string, integer, boolean)
    • Complex types (arrays, objects)
    • Date/time formats
    • Enumerations
    • Custom formats

Schema Documentation Format:

### User Object

| Field | Type | Nullable | Description |
|-------|------|----------|-------------|
| id | string | No | Unique user identifier |
| name | string | No | User's full name |
| email | string | No | User's email address |
| created_at | string (ISO 8601) | No | Account creation timestamp |
| updated_at | string (ISO 8601) | Yes | Last update timestamp |

**Example**
\`\`\`json
{
  "id": "user_123",
  "name": "John Doe",
  "email": "john@example.com",
  "created_at": "2026-05-26T12:00:00Z"
}
\`\`\`

Step 5: Code Examples and SDKs

Purpose: Provide working code examples in multiple languages.

Code Example Guidelines:

  1. Language Coverage:

    • JavaScript/Node.js (web developers)
    • Python (data scientists, backend)
    • Java/TypeScript (enterprise)
    • cURL (testing/debugging)
    • Popular frameworks (React, Vue, Django)
  2. Example Quality:

    • Complete, runnable examples
    • Include error handling
    • Show authentication
    • Display response handling
    • Add comments explaining key steps
  3. SDK Documentation (if applicable):

    • Installation instructions
    • Class/method documentation
    • Usage examples
    • Configuration options
    • Error handling patterns

Code Example Template:

import os
import requests

API_KEY = os.environ["API_KEY"]  # never hardcode credentials
BASE_URL = "https://api.example.com/v1"

def get_users(limit=20, offset=0):
    """
    Retrieve a list of users.
    
    Args:
        limit: Number of results (1-100)
        offset: Pagination offset
    
    Returns:
        List of users
    
    Raises:
        ValueError: If parameters are invalid
        requests.HTTPError: If API request fails
    """
    url = f"{BASE_URL}/users"
    headers = {"Authorization": f"Bearer {API_KEY}"}
    params = {"limit": limit, "offset": offset}
    
    try:
        response = requests.get(url, headers=headers, params=params)
        response.raise_for_status()
        return response.json()
    except requests.exceptions.HTTPError as e:
        print(f"API Error: {e}")
        raise

Step 6: Error Handling and Troubleshooting

Purpose: Help developers handle errors effectively.

Error Documentation:

  1. HTTP Status Codes:

    • Success codes (200, 201, 204)
    • Client errors (400, 401, 403, 404, 429)
    • Server errors (500, 502, 503)
    • Status code meanings and remedies
  2. Error Response Format:

    • Error response structure
    • Error codes and types
    • Error messages
    • Debug information
  3. Common Issues:

    • Authentication failures
    • Rate limiting
    • Invalid parameters
    • Network errors
    • Timeout issues
  4. Troubleshooting Guides:

    • Step-by-step problem resolution
    • Debugging tips
    • Contact support info
    • Community resources

Error Documentation Template:

## Error Handling

### HTTP Status Codes

| Status | Meaning | Resolution |
|--------|---------|------------|
| 400 | Bad Request | Fix request parameters |
| 401 | Unauthorized | Check API key |
| 403 | Forbidden | Insufficient permissions |
| 404 | Not Found | Verify endpoint URL |
| 429 | Rate Limit | Wait and retry |

### Error Response Format
\`\`\`json
{
  "error": {
    "code": "INVALID_PARAMETER",
    "message": "Invalid value for 'limit' parameter",
    "details": {
      "parameter": "limit",
      "value": 999,
      "constraint": "1-100"
    }
  }
}
\`\`\`

### Common Issues

**Issue**: "401 Unauthorized"
**Solution**: Verify API key is correct and not expired

**Issue**: "429 Rate Limit Exceeded"
**Solution**: Implement exponential backoff, wait before retry

Step 7: Versioning and Changelog

Purpose: Document API evolution and breaking changes.

Versioning Strategy:

  1. Semantic Versioning:

    • MAJOR: Breaking changes
    • MINOR: New features, backward compatible
    • PATCH: Bug fixes
  2. URL Versioning:

    • Include version in URL path (/v1/, /v2/)
    • Document deprecation timeline
    • Provide migration guides
  3. Header Versioning:

    • Use Accept header for versioning
    • Document supported versions
    • Default version behavior

Changelog Format:

## Changelog

### [Version] - [Date]

#### Added
- New endpoint description
- New parameter description

#### Changed
- Modified endpoint behavior
- Updated response format

#### Deprecated
- Endpoint deprecation notice

#### Removed
- Removed endpoint (include migration guide)

Output Format

For REST API Documentation:

  • Complete API reference with all endpoints
  • Authentication and security guide
  • Request/response schema documentation
  • Code examples in multiple languages
  • Error handling and troubleshooting guide

For GraphQL API Documentation:

  • Schema documentation with types
  • Query and mutation reference
  • Subscription documentation
  • Example queries and mutations
  • Integration guides

For SDK Documentation:

  • Installation and setup guide
  • Class/method reference
  • Usage examples and patterns
  • Configuration options
  • Error handling patterns

Best Practices

Documentation Quality: ✅ Do:

  • Start with quick start for fast onboarding
  • Provide working code examples
  • Include error handling in examples
  • Keep documentation up-to-date with API changes
  • Use consistent formatting and structure
  • Include pagination and filtering info
  • Document rate limits and throttling

❌ Don't:

  • Skip authentication documentation
  • Use placeholder code that doesn't work
  • Forget to document error responses
  • Make examples too complex
  • Ignore edge cases and errors
  • Leave documentation outdated
  • Use jargon without explanation

Developer Experience: ✅ Do:

  • Write for busy developers
  • Assume no prior API knowledge
  • Provide copy-paste examples
  • Include common workflows
  • Link to related resources
  • Make content searchable
  • Support multiple languages

❌ Don't:

  • Write like academic textbook
  • Assume readers know your API
  • Make examples too simple (unrealistic)
  • Skip error scenarios
  • Leave out important context
  • Make content hard to navigate
  • Focus on one language only

Common Mistakes to Avoid

Documentation Anti-Patterns:

  1. Outdated Docs: Documentation not synced with API changes
  2. Missing Examples: Text-only without working code
  3. Poor Search: Hard to find specific endpoints or info
  4. Inconsistent Structure: Different formats across sections
  5. Missing Context: Not explaining "why" or "when to use"

Content Anti-Patterns:

  1. Assumptions: Assuming prior knowledge
  2. Ambiguity: Unclear parameter descriptions
  3. Incomplete Info: Missing required parameters
  4. No Error Handling: Examples without error cases
  5. Dead Links: References to non-existent resources

Related Skills

  • documentation-writing: For general documentation principles
  • technical-writing: For technical communication best practices
  • code-review: For validating code examples
  • developer-experience: For overall developer portal design

API Documentation Tools

Static Site Generators:

  • Docusaurus: Facebook's React-based doc generator
  • MkDocs: Python-based static site generator
  • VuePress: Vue-powered documentation generator
  • GitBook: Book-style documentation platform

API Documentation Tools:

  • Swagger/OpenAPI: REST API specification
  • Postman: API documentation and testing
  • Insomnia: REST client and documentation
  • Stoplight: API design and documentation

Interactive API Explorers:

  • Swagger UI: OpenAPI specification visualization
  • ReDoc: Beautiful OpenAPI documentation
  • GraphQL Playground: Interactive GraphQL explorer
  • Postman: API testing and documentation

Code Examples Tools:

  • Snippets: Code snippet management
  • Carbon: Beautiful code screenshots
  • APIary: API design and documentation

Documentation Quality Checklist

Content Quality:

  • [ ] Quick start guide exists
  • [ ] All endpoints documented
  • [ ] Request parameters explained
  • [ ] Response schemas documented
  • [ ] Error responses documented
  • [ ] Code examples work correctly
  • [ ] Authentication documented
  • [ ] Rate limits specified
  • [ ] Pagination explained
  • [ ] Versioning strategy clear

Usability Quality:

  • [ ] Easy to navigate
  • [ ] Searchable content
  • [ ] Consistent formatting
  • [ ] Mobile-friendly
  • [ ] Print-friendly
  • [ ] Cross-browser compatible
  • [ ] Fast loading
  • [ ] Accessible (WCAG AA)

Maintainability Quality:

  • [ ] Version controlled
  • [ ] Easy to update
  • [ ] Clear change log
  • [ ] Deprecation policy defined
  • [ ] Migration guides provided
  • [ ] Feedback mechanism exists
  • [ ] Regularly reviewed and updated

OpenAPI/Swagger Specification

OpenAPI 3.0 Structure:

openapi: 3.0.0
info:
  title: API Name
  version: 1.0.0
  description: API description
servers:
  - url: https://api.example.com/v1
    description: Production server

paths:
  /users:
    get:
      summary: Get users
      description: Retrieve list of users
      parameters:
        - name: limit
          in: query
          schema:
            type: integer
          description: Number of results
      responses:
        '200':
          description: Successful response
          content:
            application/json:
              schema:
                type: object
                properties:
                  users:
                    type: array
                    items:
                      $ref: '#/components/schemas/User'

GraphQL Documentation

GraphQL Documentation Components:

  1. Schema Documentation:

    • Types (objects, interfaces, enums)
    • Queries (read operations)
    • Mutations (write operations)
    • Subscriptions (real-time)
  2. Operation Documentation:

    • Query/mutation signature
    • Parameters and arguments
    • Return type
    • Usage examples
  3. Example Queries:

    • Simple queries
    • Complex nested queries
    • Mutations with inputs
    • Subscriptions setup

GraphQL Documentation Format:

"""
User type definition
"""
type User {
  """
  Unique user identifier
  """
  id: ID!
  
  """
  User's full name
  """
  name: String!
  
  """
  User's email address
  """
  email: String!
  
  """
  Timestamp when user was created
  """
  createdAt: String!
}

"""
Query users
"""
type Query {
  """
  Get list of users
  """
  users(
    """
    Number of users to return
    """
    limit: Int = 20
    
    """
    Pagination offset
    """
    offset: Int = 0
  ): [User!]!
}

04 同仓库更多 Skills

5throck
5throck/ai-workspace-standards

i18n-layout

Text layout and encoding guidance: character encoding (UTF-8, legacy Korean code pages, BOM hazards), line endings, RTL/bidi handling, and script-specific font selection (Hangul focus). Use when: encoding is corrupted or mojibake appears, an RTL locale must render correctly, fonts must be chosen for a script, or CRLF/BOM issues surface.

★ 25
5throck
5throck/ai-workspace-standards

i18n-layout

Text layout and encoding guidance: character encoding (UTF-8, legacy Korean code pages, BOM hazards), line endings, RTL/bidi handling, and script-specific font selection (Hangul focus). Use when: encoding is corrupted or mojibake appears, an RTL locale must render correctly, fonts must be chosen for a script, or CRLF/BOM issues surface.

★ 25
5throck
5throck/ai-workspace-standards

i18n-layout

Text layout and encoding guidance: character encoding (UTF-8, legacy Korean code pages, BOM hazards), line endings, RTL/bidi handling, and script-specific font selection (Hangul focus). Use when: encoding is corrupted or mojibake appears, an RTL locale must render correctly, fonts must be chosen for a script, or CRLF/BOM issues surface.

★ 25
5throck
5throck/ai-workspace-standards

i18n-layout

Text layout and encoding guidance: character encoding (UTF-8, legacy Korean code pages, BOM hazards), line endings, RTL/bidi handling, and script-specific font selection (Hangul focus). Use when: encoding is corrupted or mojibake appears, an RTL locale must render correctly, fonts must be chosen for a script, or CRLF/BOM issues surface.

★ 25
5throck
5throck/ai-workspace-standards

i18n-layout

Text layout and encoding guidance: character encoding (UTF-8, legacy Korean code pages, BOM hazards), line endings, RTL/bidi handling, and script-specific font selection (Hangul focus). Use when: encoding is corrupted or mojibake appears, an RTL locale must render correctly, fonts must be chosen for a script, or CRLF/BOM issues surface.

★ 25
5throck
5throck/ai-workspace-standards

i18n-layout

Text layout and encoding guidance: character encoding (UTF-8, legacy Korean code pages, BOM hazards), line endings, RTL/bidi handling, and script-specific font selection (Hangul focus). Use when: encoding is corrupted or mojibake appears, an RTL locale must render correctly, fonts must be chosen for a script, or CRLF/BOM issues surface.

★ 25