API Settings and Integrations

Overview #

AI Workflow Automation plugin supports multiple API integrations for AI models, research tools, document processing, and web scraping. Each integration is securely managed and encrypted within your WordPress installation.

Supported APIs and Services #

AI Models and Language Processing #

1. OpenAI #

Purpose: Core AI capabilities and content generation

Models: GPT-4 Turbo, GPT-4, GPT-4o, etc.
Features: Text generation, content creation, analysis
Capabilities: Multi-modal inputs (text + images)
API Key Location: OpenAI API Keys
Usage: Primary AI processing, content generation, analysis
Pricing: Pay-per-token model

2. OpenRouter #

Purpose: Multi-model AI gateway

Models: Various including Claude, Gemini, Llama, Mistral, etc.
Features: Model switching, cost optimization
Capabilities: Access to multiple AI providers
API Key Location: OpenRouter Keys
Usage: Advanced AI processing, model comparison
Pricing: Variable based on model selection

Research and Information Gathering #

3. Perplexity #

Purpose: Advanced research and real-time information

Models: Sonar Small, Large, and Huge
Features: Online research, fact-checking, citation generation
Capabilities: Real-time web search integration
API Key Location: Perplexity Settings
Usage: Research tasks, data verification
Pricing: Per-query pricing model

Document Processing #

4. Llamaparse #

Purpose: Document parsing and data extraction

Features:
PDF parsing
Document OCR
Data structuring
Multi-language support
Capabilities:
Text extraction
Table detection
Layout preservation
Metadata extraction
API Key Location: Llamaparse Dashboard
Usage: Document processing, data extraction
Pricing: Per-page processing model
Supported Formats:
PDF documents
Scanned documents
Images with text
Office documents

Web Scraping and Crawling #

5. Firecrawl #

Purpose: Web content extraction

Features: Single page scraping, multi-page crawling
Capabilities: Content extraction, data mining
Output Formats: Text, HTML, Markdown
API Key Location: Firecrawl App
Usage: Web scraping, content aggregation
Pricing: Based on request volume

API Key Management #

Security Implementation #

Encryption:

   // Example encryption process
   $encrypted_key = WP_AI_Workflows_Encryption::encrypt($api_key);

AES-256-CBC encryption
Secure key storage in WordPress database
Salt-based key derivation

Storage Structure:

   [
       'openai_api_key' => 'encrypted_key_string',
       'perplexity_api_key' => 'encrypted_key_string',
       'openrouter_api_key' => 'encrypted_key_string',
       'firecrawl_api_key' => 'encrypted_key_string',
       'llamaparse_api_key' => 'encrypted_key_string'
   ]

Access Control:

Administrator-only access
Key masking in interface
Automatic validation

Setting Up API Keys #

Configuration Process #

Access Settings:

   WordPress Admin → AI Workflow Automation → Settings → API Configuration

Model Selection:

Enable/disable specific services
Configure model preferences
Set default parameters

API Key Entry:

Input API keys
Validation checks
Connection testing

Service-Specific Configurations #

OpenAI Setup #

Rate Limits: 
- 3 RPM (Free Trial)
- 3,000 RPM (Paid)
Max Tokens: 4096
Supported Models: GPT-4, GPT-3.5

Llamaparse Setup #

Rate Limits:
- 60 documents/hour (Standard)
- 300 documents/hour (Enterprise)
Document Size: Up to 100MB
Supported Languages: 50+
OCR Capability: Yes

Perplexity Setup #

Rate Limits:
- 60 RPM (Standard)
- 120 RPM (Enterprise)
Citation Support: Yes
Real-time Search: Enabled

OpenRouter Setup #

Rate Limits: Model-dependent
Model Switching: Enabled
Cost Tracking: Available
Usage Analytics: Included

Firecrawl Setup #

Rate Limits:
- 60 RPM (Basic)
- 300 RPM (Premium)
Max Pages: 1000/crawl
Javascript Rendering: Optional

Usage Monitoring and Optimization #

Usage Tracking #

Per-Service Monitoring:

API call counts
Cost tracking
Error rates
Response times

Usage Alerts:

Quota notifications
Error alerts
Cost thresholds
Performance warnings

Optimization Features #

Cost Management:

Model selection optimization
Batch processing
Cache utilization
Request consolidation

Performance Tuning:

Response caching
Rate limit management
Concurrent processing
Error handling

Best Practices #

Security Guidelines #

API Key Management:

Regular key rotation
Environment-based keys
Access logging
Secure storage

Usage Patterns:

Request batching
Error handling
Rate limit compliance
Response validation

Performance Optimization #

Request Management:

Queue implementation
Batch processing
Cache utilization
Load balancing

Error Handling:

Retry logic
Fallback options
Error logging
User notifications

Troubleshooting #

Common Issues and Solutions #

Issue	Service	Solution
Rate Limiting	All	Implement request queuing
Authentication	All	Verify API key format
Timeout	Llamaparse	Adjust document size
Parsing Error	Llamaparse	Check file format
Model Error	OpenAI	Verify model availability

Error Response Handling #

try {
    // API call
} catch (APIException $e) {
    // Log error
    WP_AI_Workflows_Utilities::debug_log("API Error", "error", [
        'service' => $service,
        'error' => $e->getMessage()
    ]);
}

Support and Resources #

Documentation Links #

Getting Help #

Plugin support forums
Service provider support
Community resources
Technical documentation

Security Advisory #

Key Protection Measures #

Storage Security:

Encrypted storage
Secure transmission
Access control
Key rotation

Access Control:

Role-based access
Activity logging
Session management
IP restrictions

Monitoring:

Usage tracking
Error logging
Access attempts
System alerts

Updates and Maintenance #

Regular Tasks #

API Maintenance:

Key validation
Usage review
Performance monitoring
Security updates

System Requirements:

WordPress 6.0+
PHP 7.2+
MySQL 5.6+
HTTPS enabled

Remember to regularly review and update your API configurations to ensure optimal security and performance. Always follow the specific guidelines provided by each service provider for best results.