Skip to main content

​
Overview

The Splox integration provides built-in tools and utilities for workflow automation, data management, and agent orchestration within the Splox platform.

Workflow Builder

Create and manage workflows and agents programmatically

Data Storage

Key-value store and vector database operations

Utilities

Date/time, crypto, conversion, and API tools

LLM Operations

Model management and token generation

​
Setup

The Splox integration is automatically available in all workflows - no additional setup required.
Built-in Integration: All Splox tools are available by default in every workflow.

​
Available Actions

​
Workflow Builder

Create a new workflow in your workspace.Parameters:
  • name - Workflow name
  • description - Workflow description (optional)
  • nodes - Array of node configurations
  • edges - Array of edge connections
  • settings - Workflow settings (optional)
Returns:
  • workflow_id - Created workflow ID
  • name - Workflow name
  • created_at - Creation timestamp
Retrieve workflow details and configuration.Parameters:
  • workflow_id - Workflow ID to retrieve
Returns:
  • id - Workflow ID
  • name - Workflow name
  • description - Workflow description
  • nodes - Array of nodes
  • edges - Array of edges
  • settings - Workflow settings
  • created_at - Creation date
  • updated_at - Last update date
Delete nodes from a workflow.Parameters:
  • workflow_id - Target workflow ID
  • node_ids - Array of node IDs to delete
Returns:
  • success - Deletion success status
  • deleted_count - Number of nodes deleted
Create a new AI agent.Parameters:
  • name - Agent name
  • description - Agent description
  • llm_config - LLM configuration
    • provider - AI provider (openai, anthropic, etc.)
    • model - Model name
    • temperature - Temperature setting
    • max_tokens - Max token limit
  • system_prompt - Agent system prompt
  • tools - Array of tool IDs to attach
Returns:
  • agent_id - Created agent ID
  • name - Agent name
  • created_at - Creation timestamp
Update an existing agent’s configuration.Parameters:
  • agent_id - Agent ID to update
  • name - New name (optional)
  • description - New description (optional)
  • llm_config - Updated LLM config (optional)
  • system_prompt - New system prompt (optional)
Returns:
  • agent_id - Updated agent ID
  • updated_at - Update timestamp
Attach tools to an agent.Parameters:
  • agent_id - Target agent ID
  • tool_ids - Array of tool IDs to add
Returns:
  • agent_id - Agent ID
  • tools - Updated tool list
  • tool_count - Total number of tools
Connect multiple agents for collaboration.Parameters:
  • agent_ids - Array of agent IDs to connect
  • connection_type - Connection type (sequential, parallel, hierarchical)
  • communication_protocol - How agents communicate
Returns:
  • connection_id - Created connection ID
  • agent_count - Number of connected agents
Create a multi-agent system with multiple collaborating agents.Parameters:
  • name - System name
  • description - System description
  • agents - Array of agent configurations
  • orchestration_type - How agents work together
    • sequential - One after another
    • parallel - All at once
    • hierarchical - Manager and workers
    • debate - Agents debate and vote
  • communication - Communication settings
Returns:
  • system_id - Created system ID
  • agent_ids - Array of created agent IDs
  • orchestration - Orchestration configuration
Search for available integrations in Splox.Parameters:
  • query - Search query
  • category - Filter by category (llm, social, productivity, etc.)
  • limit - Number of results
Returns:
  • integrations - Array of matching integrations
    • id - Integration ID
    • name - Integration name
    • description - Integration description
    • category - Integration category
    • actions - Available actions
List Model Context Protocol (MCP) servers configured by the user.Returns:
  • mcps - Array of user MCP servers
    • id - MCP ID
    • name - MCP name
    • url - MCP endpoint
    • status - Connection status
    • tools - Available tools
List publicly available MCP templates.Parameters:
  • category - Filter by category (optional)
  • limit - Number of results
Returns:
  • templates - Array of MCP templates
    • id - Template ID
    • name - Template name
    • description - Template description
    • author - Template creator
    • tools - Included tools
    • install_url - Installation URL

​
Key-Value Store

Retrieve a value from the key-value store.Parameters:
  • key - Key to retrieve
  • workspace_id - Workspace ID (optional)
Returns:
  • value - Stored value
  • exists - Whether key exists
Store a value in the key-value store.Parameters:
  • key - Key to store
  • value - Value to store
  • workspace_id - Workspace ID (optional)
  • ttl - Time to live in seconds (optional)
Returns:
  • success - Operation success status
Delete a key from the key-value store.Parameters:
  • key - Key to delete
  • workspace_id - Workspace ID (optional)
Returns:
  • success - Deletion success status
Append to a string value in the store.Parameters:
  • key - Key to append to
  • value - Value to append
  • workspace_id - Workspace ID (optional)
Returns:
  • new_value - Updated value
  • length - New length
Append items to an array in the store.Parameters:
  • key - Key to append to
  • items - Array of items to append
  • workspace_id - Workspace ID (optional)
Returns:
  • new_array - Updated array
  • length - New array length
Increment a numeric value in the store.Parameters:
  • key - Key to increment
  • increment_by - Amount to increment (default: 1)
  • workspace_id - Workspace ID (optional)
Returns:
  • new_value - Updated value
Decrement a numeric value in the store.Parameters:
  • key - Key to decrement
  • decrement_by - Amount to decrement (default: 1)
  • workspace_id - Workspace ID (optional)
Returns:
  • new_value - Updated value

​
Vector Store

Insert documents into the vector store.Parameters:
  • documents - Array of documents to insert
    • id - Document ID (optional)
    • text - Document text content
    • metadata - Document metadata (optional)
  • collection_name - Collection name (optional)
Returns:
  • inserted_ids - Array of inserted document IDs
  • count - Number of documents inserted
Rerank search results using a cross-encoder model.Parameters:
  • query - Search query text
  • documents - Array of documents to rerank
  • top_k - Number of top results to return
Returns:
  • reranked_results - Array of reranked documents with scores

​
Date & Time

Get the current date and time.Parameters:
  • timezone - Timezone (optional, default: UTC)
  • format - Output format (optional)
Returns:
  • datetime - Current date/time
  • timestamp - Unix timestamp
  • timezone - Used timezone
Format a date/time into a specific format.Parameters:
  • datetime - Date/time to format
  • format - Desired format (e.g., “YYYY-MM-DD HH:mm:ss”)
  • timezone - Timezone (optional)
Returns:
  • formatted - Formatted date/time string
Add a duration to a date/time.Parameters:
  • datetime - Starting date/time
  • years - Years to add (optional)
  • months - Months to add (optional)
  • days - Days to add (optional)
  • hours - Hours to add (optional)
  • minutes - Minutes to add (optional)
  • seconds - Seconds to add (optional)
Returns:
  • result - New date/time
Subtract a duration from a date/time.Parameters:
  • datetime - Starting date/time
  • years - Years to subtract (optional)
  • months - Months to subtract (optional)
  • days - Days to subtract (optional)
  • hours - Hours to subtract (optional)
  • minutes - Minutes to subtract (optional)
  • seconds - Seconds to subtract (optional)
Returns:
  • result - New date/time
Extract a specific part from a date/time.Parameters:
  • datetime - Date/time to extract from
  • part - Part to extract (year, month, day, hour, minute, second, weekday)
Returns:
  • value - Extracted value
Calculate the difference between two date/times.Parameters:
  • start_datetime - Start date/time
  • end_datetime - End date/time
  • unit - Unit of measurement (days, hours, minutes, seconds)
Returns:
  • difference - Time difference in specified unit
Round a date/time to the nearest unit.Parameters:
  • datetime - Date/time to round
  • unit - Unit to round to (hour, day, week, month, year)
  • direction - Rounding direction (up, down, nearest)
Returns:
  • rounded - Rounded date/time

​
Crypto

Generate random cryptographic data.Parameters:
  • type - Type to generate (uuid, random_bytes, random_string)
  • length - Length of output (for bytes/string)
  • charset - Character set for strings (optional)
Returns:
  • value - Generated cryptographic data
Generate a hash of input data.Parameters:
  • data - Data to hash
  • algorithm - Hash algorithm (md5, sha1, sha256, sha512)
  • encoding - Output encoding (hex, base64)
Returns:
  • hash - Generated hash
  • algorithm - Used algorithm
Generate an HMAC signature.Parameters:
  • data - Data to sign
  • secret - Secret key
  • algorithm - HMAC algorithm (sha256, sha512)
  • encoding - Output encoding (hex, base64)
Returns:
  • hmac - Generated HMAC signature
Create a cryptographic signature.Parameters:
  • data - Data to sign
  • private_key - Private key for signing
  • algorithm - Signing algorithm (RSA, ECDSA)
Returns:
  • signature - Generated signature
  • algorithm - Used algorithm

​
Conversion

Convert HTML content to a PDF file.Parameters:
  • html - HTML content to convert
  • options - Conversion options (optional)
    • page_size - Page size (A4, Letter, etc.)
    • orientation - Page orientation (portrait, landscape)
    • margins - Page margins
Returns:
  • pdf_url - URL to generated PDF
  • file_size - PDF file size
Convert HTML content to an image.Parameters:
  • html - HTML content to convert
  • format - Image format (png, jpeg)
  • width - Image width (optional)
  • height - Image height (optional)
Returns:
  • image_url - URL to generated image
  • dimensions - Image dimensions
Merge multiple PDF files into one.Parameters:
  • pdf_urls - Array of PDF file URLs to merge
  • order - Order of merging (optional)
Returns:
  • merged_pdf_url - URL to merged PDF
  • page_count - Total pages in merged PDF
Capture a screenshot of a URL.Parameters:
  • url - URL to capture
  • format - Image format (png, jpeg)
  • full_page - Capture full page or viewport
  • width - Viewport width (optional)
  • height - Viewport height (optional)
Returns:
  • screenshot_url - URL to screenshot
  • dimensions - Screenshot dimensions
Convert audio between different formats.Parameters:
  • audio_url - URL of audio file to convert
  • target_format - Target format (mp3, wav, ogg, etc.)
  • bitrate - Audio bitrate (optional)
  • sample_rate - Sample rate (optional)
Returns:
  • converted_url - URL to converted audio
  • duration - Audio duration
  • format - Output format

​
LLM Operations

List all available LLM models.Parameters:
  • provider - Filter by provider (optional)
  • category - Filter by category (optional)
Returns:
  • models - Array of available models
    • id - Model ID
    • name - Model name
    • provider - Model provider
    • capabilities - Model capabilities
    • context_window - Context window size
Generate a temporary API token for chat completions.Parameters:
  • duration - Token validity duration (optional)
  • permissions - Token permissions (optional)
Returns:
  • token - Generated API token
  • expires_at - Token expiration time

​
Internal Workflow Communication

Send a message in the internal workflow chat.Parameters:
  • chat_id - Workflow chat ID
  • message - Message content
  • metadata - Optional metadata
Returns:
  • message_id - Sent message ID
  • timestamp - Message timestamp
Mark a message as complete in workflow chat.Parameters:
  • chat_id - Workflow chat ID
  • message_id - Message to mark complete
Returns:
  • success - Completion status

​
Utility

Make HTTP API requests to external services.Parameters:
  • url - API endpoint URL
  • method - HTTP method (GET, POST, PUT, DELETE, PATCH)
  • headers - Request headers (optional)
  • body - Request body (optional)
  • params - Query parameters (optional)
  • auth - Authentication details (optional)
Returns:
  • status_code - HTTP status code
  • body - Response body
  • headers - Response headers
Check the status of a workflow execution.Parameters:
  • workflow_request_id - Workflow execution ID
Returns:
  • status - Execution status
  • progress - Execution progress
  • result - Execution result (if complete)
  • error - Error message (if failed)

​
Use Cases

Use KV Store to maintain state between workflow executions and share data across workflows.
Create dynamic workflows that generate and modify other workflows programmatically.
Build complex agent systems where multiple AI agents collaborate on tasks.
Convert and merge documents, generate PDFs from HTML, capture screenshots.
Process dates, generate hashes, make API calls to external services.

​
Best Practices

Data Management

Use namespaced keys in KV Store to avoid conflicts

Vector Search

Organize documents in collections for better search performance

Agent Design

Create specialized agents with focused roles and clear responsibilities

Error Handling

Always handle API request failures and timeouts gracefully

Security

Use secure key generation for sensitive operations

Performance

Cache frequently accessed KV Store values

​
Limitations

  • KV Store: 10MB max value size per key
  • Vector Store: 100,000 documents per collection
  • Workflow Size: 1000 nodes per workflow
  • Agent Systems: 50 agents per multi-agent system
  • API Requests: 60 requests per minute
  • File Conversions: 50MB max file size
Resource Limits: Monitor usage to stay within quotas. Contact support for higher limits.

​
Next Steps

Multi-Agent Guide

Learn about multi-agent systems

Subflow Node

Create reusable subflows

API Reference

Trigger workflows via API

Agent Documentation

Single agent workflows