Bank Detail AI Agent

Agent Description:

The Bank Detail AI Agent securely automates the collection and verification of customer banking information. It fetches records via API, ensuring the data is clean, properly formatted, and that sensitive details are masked. The agent then cross-verifies account numbers and IFSC codes against secure banking APIs. Finally, it updates the status to "Verified" or "Mismatch," logs discrepancies, and generates a summary report for compliance.

Purpose and Components
  • Purpose: To automate the secure collection, formatting, and verification of customer banking information (like account numbers and IFSC codes) to ensure data accuracy and compliance.
  • Components:

    • Data Fetcher: An agent to securely retrieve customer banking records from an API source.

    • Validation Engine: An agent that cross-verifies the fetched data against banking APIs and logs the verification status.

    • API Connector (GET): A tool to fetch the raw customer banking data.

Supported Capabilities

  • Initiating a secure GET API call to retrieve customer banking records.

  • Ensuring all required fields (account number, IFSC, bank name) are present and correctly formatted.

  • Masking or encrypting sensitive fields during processing.

  • Forwarding cleaned and structured data for validation.

  • Cross-verifying ifsc_code and account_number for each customer record (presumably using another secure API, though not explicitly listed as a tool for the second agent).

  • Updating a verification_status as "Verified" or "Mismatch".

  • Logging discrepancies and generating a verification summary report.

  • Sending a confirmation message upon successful verification.

LLM Used

  • GPT_4O_MINI

    Note: To learn more about the LLM and to modify its behavior, refer to the Configuring LLM settings section.

Sub-Agents

1. Data Fetcher

  • Role: Data Collector

  • Scope: Fetches customer banking details using GET API requests and OCR results. Cleans and formats records for validation.

  • Description: This agent starts the process by making a secure GET API call to retrieve all customer banking records. It validates that all required fields are present and correctly formatted, and ensures sensitive data is masked. It then forwards the cleaned dataset to the Validation Engine.

  • LLM Used: Default (Inherits from parent agent).

  • Tools Used: Request - Get

2. Validation Engine

  • Role: Data Verifier

  • Scope: Validates accuracy of fetched banking data using secure internal/external APIs and updates verification status.

  • Description: This agent receives the cleaned data from the Data Fetcher. It performs the core verification by cross-checking the ifsc_code and account_number against a secure banking API. It then updates the record's status to "Verified" or "Mismatch," logs any errors, and generates a final summary report.

  • LLM Used: Default (Inherits from parent agent).

  • Tools Used: None

Tools Used:

  • Request - Get Tool: Fetches the raw customer banking records from a remote JSON endpoint.

Note: For details on modifying the Tools, refer Tools Library section.
Agent Workflow Behavior Summary

  1. The Data Fetcher (start node) is triggered and uses the Request - Get tool to retrieve a dataset of customer banking records.

  2. It validates the format of this data and ensures it is clean and secure (masked/encrypted).

  3. The structured dataset is then passed to the Validation Engine.

  4. The Validation Engine (end node) cross-verifies the account number and IFSC code for each record.

  5. It generates a final report, updating the status of each record to "Verified" or "Mismatch" and sends a confirmation.

Sample Questions:

  • Show me all verified customer bank details.

  • What is the IFSC code of <customer_name> account?

  • Identify customers whose verification status is ‘Pending Validation’