Docs / Strand / workflow-editor/node-types

Node Types Overview

Complete overview of all available node types in Strand workflows.

Selecting a node opens the Node Inspector, where its type-specific configuration lives.

Built-in Node Types

Logic (logic)

Multi-mode control flow node for filtering, branching, and iteration.

Modes:

• Filter - Allow/block events based on Python conditions
• If/Else - Branch to "if" or "else" outputs based on Jinja2 condition
• Foreach - Iterate over arrays with batched execution and result aggregation

See: Logic Node Documentation

Transform (transform.map)

Transforms event data by mapping fields. Creates new payloads from existing data.

Example:

{
  "full_name": "{{ payload.first_name }} {{ payload.last_name }}",
  "email": "{{ payload.email }}"
}

See: Transform Node Documentation

HTTP Request (http.request)

Makes HTTP requests to external APIs. Supports templating in URLs, headers, body, query params, and path params. Configurable retry with exponential/linear/fixed backoff.

See: HTTP Request Node Documentation

Flow Call (flow.call)

Executes another workflow as a step in the current workflow. Supports passing custom data with Jinja2 templating. Max 10 levels of nesting.

See: Flow Call Node Documentation

Function Call (function.call)

Executes a reusable Python function from your function library.

See: Function Call Node Documentation

Python Snippet (python.snippet)

Run custom Python code in a secure sandbox. Access payload, meta, and set output.

See: Python Snippet Node Documentation

Delay (delay.ms)

Adds a delay (in milliseconds) before continuing to the next node.

See: Delay Node Documentation

Print (print)

Logs a message for debugging. Supports Jinja2 templating. Automatically redacts vault values.

See: Print Node Documentation

Approval (approval)

Pauses workflow execution until a human approves or rejects. Supports Jinja2 templating in the approval message, optional reviewer instructions, external notifications via connectors, and approve-from-link with signed URLs.

See: Approval Node Documentation

Comment (comment)

A no-op documentation node for annotating your workflow canvas. The event passes through unchanged; Comment nodes do not modify data. Use them to label sections, explain logic, or leave notes for collaborators.

Configuration:

• Text: Comment text content
• Font Size: Text size (0.75rem, 0.875rem, 1rem, 1.25rem)

Platform Nodes

Platform nodes are always available in the node selector; they don't require a saved connector instance. See Connector Nodes for full field and output details.

Surface: Scan Payload (platform.surface.scan)

Scans payload content for threats using Surface. Required field: content. Optional: reject_on, profile_id, label, timeout. Scan results are written flat onto the payload (threat_level, score, iocs, primary_threat, recommended_action, safe).

Contact: Send Message (platform.contact.send)

Sends a message via Contact using an entity API key. Required field: api_key. Optional: dest, tags, data, timeout.

Contact: Send File (platform.contact.send_file)

Sends a file into Contact using an entity API key (Contact file transfer). Required fields: api_key and content_base64 (the file bytes, base64-encoded; it chains directly from an upstream Contact: Fetch File or HTTP node). Set exactly one delivery target: dest (a recipient entity, for direct/group/cross-account delivery), tags (route to subscribing Strand workflows/connectors), or kind: clip (motion gallery, no recipient). Optional: file_name, content_type, meta (≤8 KB), timeout. The sending entity needs the entity:SendFiles permission and pays storage until the file is consumed. On success the node adds transfer_id, sha256, size, and mode to the payload; any Contact rejection (missing permission, over quota/size limit, or a file flagged by Surface) fails the node.

Contact: Update State (platform.contact.update_state)

Updates an entity's state table in Contact. Required fields: api_key, entity, state. Optional: replace (default merges; true replaces the whole table), timeout.

Connector Nodes

Connector nodes use saved connectors to interact with external services. They are dynamically created from your configured connectors.

Available Connector Types

See: Connector Nodes Documentation

Node Selection Guide

Choosing the Right Node

- Logic (Filter) - Allow/block events conditionally

- Logic (If/Else) - Branch into two paths

- Logic (Foreach) - Process each item in an array

- Transform - Simple data mapping and reshaping

- HTTP Request - Direct API calls (with or without connectors)

- AI / LLM Connector - AI text generation, classification, extraction

- Python Snippet - Complex data processing and calculations

- Function Call - Reusable logic shared across workflows

- Flow Call - Reusable workflow components

- Approval - Human review checkpoint before proceeding

- Delay - Rate limiting or scheduling

- Print - Debugging

- Comment - Annotate and document your workflow

Related

• Creating Workflows - Step-by-step workflow creation
• Connecting Nodes - How to connect nodes
• Conditional Logic - Branching, filtering, and iteration with the Logic node
• Templating Overview - Data access and transformation
• External Approval Links - Approve/reject from email or Slack

← Back to Strand docs