AI Actions

LLM-powered intelligence nodes for reasoning and generation.

Provider Configuration

AI actions use your configured LLM provider:

• Ollama (local, private)
• OpenAI (GPT-4, GPT-3.5)
• Anthropic (Claude)
• Groq (fast inference)

Configure in Settings → AI Provider.

Available Actions

ai.generate

Free-form text generation.

Input	Type	Description
prompt	String	The instruction/question
context	String (optional)	Additional context
system	String (optional)	System prompt

Output	Type	Description
response	String	Generated text

ai.summarize

Condense long text.

Input	Type	Description
text	String	Text to summarize
length	String	"short", "medium", "long"

ai.classify

Categorize input into predefined classes.

Input	Type	Description
text	String	Text to classify
categories	Array	List of possible categories

Output	Type	Description
category	String	Matched category
confidence	Number	0-1 confidence score

ai.extract

Pull structured data from text.

Input	Type	Description
text	String	Source text
schema	Object	JSON schema to extract

Output	Type	Description
data	Object	Extracted structured data

ai.sentiment

Analyze emotional tone.

Input	Type	Description
text	String	Text to analyze

Output	Type	Description
sentiment	String	"positive", "negative", "neutral"
score	Number	-1 to 1

Blueprint Examples

Smart Email Triage

[Email Received]
        │
        ▼
[AI Classify: categories = ["urgent", "fyi", "spam"]]
        │
        ├── urgent → [Notification]
        ├── fyi → [Archive]
        └── spam → [Delete]

Meeting Notes

[Voice Recording]
        │
        ▼
[AI Summarize: length = "short"]
        │
        ▼
[Save to Notes]

Best Practices

1. Be specific: Clear prompts get better results
2. Use context: Pass relevant memory/data for better understanding
3. Validate output: AI can make mistakes—verify critical actions