AI Business Analyst
No ratings yet
645Multi-Agent Requirements Architect & QA Auditor, which automates the process of transforming raw, disorganized business feedback into structured, development-ready product backlog requirements.
Operations
Setup steps
⚙️ How to Configure and Use This Template
Follow these quick steps to get this multi-agent requirements pipeline running in your workspace:
-
Import the Workflow
- Click "Open in Dify" or click "Use Template" to clone this workflow directly into your workspace studio.
-
Configure Your LLM Models
- Open the first LLM node (Requirements Lead). Ensure a model is selected (e.g., gpt-4o-mini, Claude 3.5 Haiku).
- Open the LLM 2 node (Auditor Agent) and confirm it is using your desired model as well.
-
Verify the Pipeline Variables
- Double-check that the first LLM node's prompt reads from your node's user input variable.
- Double-check that LLM 2 (Auditor Agent) is configured to read the output from the first LLM node. This ensures the auditor reviews the draft instead of the raw input.
- Open the final orange Output 2 node and verify that its output variable is cleanly mapped to the string coming directly from LLM 2.
-
Publish and Run a Test
- Click the Publish button in the upper right-hand corner of your master canvas.
- Click the Run (▶) button to launch the live testing preview panel.
- Paste any messy, disorganized user request or rough product feature description into the text block and hit execute.
- Navigate to the Result or Last Run tab to view your beautifully audited, development-ready Jira User Story and Acceptance Criteria!
Give a rate
TYPE
Workflow
PLUGINS
Openai
Dify Compatibility
1.14.0
SUPPORTED LANGUAGES
English
LAST UPDATED
Sep 10, 2026
SHARE WITH