Quickstart: RAG-Grounded Code Generation

Run a retrieval-grounded code generation workflow that cites real attune APIs and workflow names — never invented ones.

from workflows.rag_code_gen import RagCodeGenWorkflow

workflow = RagCodeGenWorkflow()
result = workflow.execute()
print(result)

Prerequisites

• The project is cloned and installed locally
• You can import from workflows.rag_code_gen

Steps

1. Import and instantiate

from workflows.rag_code_gen import RagCodeGenWorkflow

workflow = RagCodeGenWorkflow()

2. Execute the workflow

Call execute() to run retrieval and grounded generation:

result = workflow.execute()

3. Inspect the result

Print the returned WorkflowResult to confirm grounded output was produced:

print(result)

You should see a WorkflowResult containing generated code and citations drawn from retrieved attune-help context. Any API names, workflow names, or CLI commands in the output will reference real attune source files.

Next:

Read the concept page for RAG grounding to understand how retrieved passages are injected into the prompt and why the system prompt rejects invented attune features.

Unresolved references

Auto-generated by attune-author fact-check. Review and either fix the source code, fix this doc, or add an override.