guardrails

Overview ¶

This example demonstrates output validation guardrails:

• Task-level guardrails with retry — LengthGuardrail + ContainsGuardrail
• Agent-level guardrails — BannedWordsGuardrail
• JSONSchemaGuardrail for structured output

Prerequisites:

• An LLM API key (OpenAI or Ollama)

Usage (OpenAI):

OPENAI_API_KEY=sk-... go run main.go

Usage (Ollama):

OLLAMA_BASE_URL=http://localhost:11434 OLLAMA_MODEL=gemma3:4b go run main.go