Actions10
- Orchestration Actions
- Vector Store Actions
Orchestration → Guarded Generate
AI-generatedSummary
Generate text from a connected AI model with optional system prompt, temperature, max tokens, and guarded mode flag included in output.
Inputs
- prompt (required) — Prompt or query text
- systemPrompt — Optional system instruction
- temperature — Sampling temperature
- maxTokens — Maximum number of output tokens
- guardedMode — When enabled, guarded mode flag is included in the output for downstream reliability checks
Output shape
object with resource, operation, model, text, data, usage, guardedMode, and request fields
A connected AI model must be attached to the Chat Model input; otherwise the node throws an error.