AI-Powered RFP Response Generator Agent

Unlock Rapid RFP Turnaround with this AI Agent

This n8n workflow acts as an intelligent AI Agent designed to dramatically speed up and streamline your Request for Proposal (RFP) response process. It's triggered by a webhook where you submit your RFP (PDF) and some metadata. The agent then takes over:

1. RFP Ingestion & Preprocessing: The PDF is received, and its text content is extracted.
2. AI-Powered Question Extraction: Using a Langchain LLM chain powered by an OpenAI chat model, the agent intelligently identifies and extracts all questions from the RFP text. This ability ensures robust parsing across varying RFP formats.
3. AI-Driven Answer Generation: For each extracted question, the agent consults an OpenAI Assistant that you've previously trained with your company's specific knowledge base (e.g., product documentation, past Q&As, marketing materials). This ensures the agent's ability to provide contextually relevant and accurate answers.
4. Response Compilation & Document Assembly: A new Google Document is created for the RFP response. Metadata and each question-answer pair are systematically added to this document.
5. Team Notification: Once the draft response is complete, notifications are sent via Gmail and Slack, including a link to the generated Google Doc.

Key Features & Benefits

• Automated RFP Ingestion: Handles PDF documents submitted via a secure webhook.
• Intelligent Question Extraction Ability: Leverages Langchain and OpenAI's LLMs (e.g., GPT models) to accurately parse questions from diverse RFP layouts.
• Contextual Answer Generation Ability: Utilizes a custom OpenAI Assistant fed with your company's unique data for tailored and precise responses.
• Automated Document Creation Ability: Generates a clean, organized RFP response in Google Docs.
• Team Alerting Ability: Keeps stakeholders informed with automated Gmail and Slack notifications.
• Significant Time Savings: Drastically reduces the manual effort involved in drafting RFP responses.
• Improved Response Quality & Consistency: Ensures answers are aligned with company knowledge and tone.
• Scalable Solution: Enables sales and proposal teams to handle a higher volume of RFPs more efficiently.

• An n8n instance (Cloud or self-hosted).
• OpenAI API Key with access to GPT models and Assistants API.
• An OpenAI Assistant created in the OpenAI platform, configured with Retrieval capabilities and uploaded with your company's knowledge base documents (e.g., product specs, FAQs, marketing brochures).
• Google Workspace account with API access enabled for Google Docs and Gmail (OAuth2 credentials configured in n8n).
• Slack workspace with API access for notifications (Bot token configured in n8n).
• A tool like Postman or cURL to make the initial POST request to the webhook trigger.

1. Download the n8n workflow JSON file.
2. Import the workflow into your n8n instance.
3. Webhook Node ('Wait for Request'): Note the Test URL. For production, use the Production URL and ensure it's secured. This endpoint expects a POST request with form-data fields: id (RFP identifier), title (RFP title), reply_to (email for notifications), and data (the RFP PDF file).
4. Google Docs Nodes ('Create new RFP Response Document', 'Add Metadata...', 'Record Question & Answer...'): Authenticate with your Google account in n8n. Optionally, update the folderId in the 'Create new RFP Response Document' node to save documents to a specific Google Drive folder.
5. OpenAI Chat Model Node (connected to 'Extract Questions From RFP'): Configure this node with your OpenAI API credentials. This model is used by the Langchain chain for question extraction.
6. 'Answer Question with Context' (OpenAI Assistant Node): Select your OpenAI API credentials. Choose 'Assistant' as the resource. Select your pre-configured OpenAI Assistant by its ID (the Assistant should have your company documents uploaded for context).
7. Gmail Node ('Send Email Notification'): Configure with your Gmail account credentials in n8n. The sendTo field is dynamically set from the webhook's reply_to input. Customize the subject and message as needed.
8. Slack Node ('Send Chat Notification'): Configure with your Slack account credentials in n8n. Select the target channel and customize the message.
9. Review all nodes for any specific placeholder values (e.g., example folder IDs, channel names) and update them to your requirements.
10. Test the workflow by sending a POST request to the webhook URL using a tool like Postman (refer to Sticky Note9 in the workflow for an example cURL command structure).
11. Activate the workflow for live RFP processing.