How to build a Closing Compliance Agent

TL;DR

This agent automates the review of loan and credit application documents, detects inconsistencies and fraud risks using AI, and routes high-risk cases for human review while logging low-risk cases for record-keeping.

What It Does:

• Ingests and processes uploaded loan application documents (including scanned files with OCR).

• Analyzes documents with an AI model trained to spot inconsistencies, fraud indicators, and risk factors.

• References a knowledge base of fraud indicators and performs web searches for up-to-date verification.

• Classifies applications as high-risk or low-risk using an AI routing node.

• Automatically notifies reviewers via email for high-risk applications.

• Logs low-risk applications to Google Drive for compliance and tracking.

Who It’s For:

• Loan officers and underwriters

• Credit risk teams

• Financial institutions and banks

• Compliance and fraud detection teams

Time to Value:

• Immediate: Upload documents and get a risk assessment, summary, and routing decision in minutes—no manual review required.

Output:

• For high-risk applications:

  • Detailed AI findings and recommendations

  • Automated email alert to the reviewer

• For low-risk applications:

  • AI summary and risk assessment

  • Record automatically created in Google Drive

Common Pain Points for Closing

• Manual review is slow, error-prone, and inconsistent

• Fraud indicators are often missed due to volume or lack of expertise

• High-risk cases may not be escalated promptly

• Record-keeping for compliance is tedious

• Difficulty in keeping up with new fraud tactics and up-to-date information

What This Agent Delivers

• Automated, consistent document analysis and risk detection

• Real-time fraud indicator referencing and web verification

• Clear, actionable summaries and recommendations

• Instant routing of high-risk cases to human reviewers

• Automated record-keeping for low-risk cases

• Reduced manual workload and faster decision-making

Step-by-Step Build (StackAI Nodes)

1) Files Node (doc-0)

What it does:

• Accepts user-uploaded files (PDFs, images, etc.).

• Extracts and processes text, including OCR for scanned documents.

Goal:

• Provide all document content for downstream AI analysis.

2) LLM Node (llm-0)

What it does:

• Analyzes the uploaded documents.

• Detects document types, checks for missing/expired/non-compliant items.

• Outputs a checklist with status symbols and a summary.

Goal:

• Automate expert-level review and checklist generation.

Instructions

You are “The Gatekeeper,” an expert in reviewing closing packages for large loans. Your job is to verify the completeness, 
  compliance, and currency of all uploaded documents

Prompt

Instructions: 

You are “The Gatekeeper,” an expert in reviewing closing packages for large loans. 



Review the uploaded documents and categorize each as one of the following: 

- Present (all required and up-to-date): Mark with ✅ 

- Missing (required but not provided): Mark with ❌ - Expired (provided but expired or non-compliant): Mark with ⚠️ 



Output the checklist in three sections, using the symbols above: 

1. Present Documents - List each present document with a ✅ and a short description. 

2. Missing Documents - List each missing or incomplete required document with a ❌ and a short description. 

3. Expired/Non-Compliant Documents - List each expired or non-compliant document with a ⚠️ and a short description. 

At the top, include a subject line: "Closing Package Review - Action Required by [due date]" Greet the user by name if available (otherwise use "Dear Borrower,"). Mention the loan number if available (otherwise leave blank). State the closing date and due dates for each section if possible. Add a "Timeline" section with all due dates. Output everything as plain text (not HTML or Markdown table). Input documents: {{doc-0.documents}}

3) Python Node (python-0)

What it does:

• Receives the LLM’s checklist output.

• Checks for the presence of ❌ or ⚠️ symbols.

• If issues are found, passes the checklist through; otherwise, outputs an empty string.

Goal:

• Ensure only problematic checklists trigger notifications.

4) Template Node (template-0)

What it does:

• Formats the checklist and review summary in markdown.

• Uses the Python node’s output, so only displays/sends the checklist if issues exist.

Goal:

• Create a user-friendly, professional report for output and email.

5) Output Node (out-0)

What it does:

• Displays the formatted checklist and review summary to the user.

Goal:

• Provide immediate, clear feedback in the app.

6) Send Email Action Node (action-0)

What it does:

• Sends an email with the checklist if issues are found.

• Uses a pre-configured Gmail connection and sends to a specified recipient.

Goal:

• Automatically alert stakeholders when action is required.

7) Sticky Note Node (stickynote_v2-0)

What it does:

• Provides a visual summary and instructions within the workflow builder.

Goal:

• Help users understand the workflow’s purpose and logic.