Automate Public Disclosure Using Amazon Q Developer CLI
I introduce a scalable way to handle Public Disclosure Requests using Amazon Q Developer CLI - fully automated via natural language in your terminal.
Arun Thangavel
Amazon Employee
Published Apr 30, 2025
Last Modified May 1, 2025
Public sector agencies often operate under strict compliance timelines, especially when it comes to responding to public disclosure requests. For customers using Amazon Connect as their contact center solution, fulfilling these requests can be labor-intensive, involving manual extraction of redacted audio and transcriptions—and further complicated when the transcriptions are stored in raw JSON format.
The current methods demand either technical scripting skills or a significant amount of manual work—posing a real bottleneck for teams that are resource-constrained.
Public sector contact centers using Amazon Connect face several roadblocks:
- Bulk Request Handling: Disclosure requests often involve hundreds of Contact IDs at once.
- Technical Expertise Gap: Staff may lack the ability to write custom scripts for data extraction and formatting.
- Manual JSON Processing: Transcriptions are available in JSON format, requiring tedious manual conversion into readable transcript text.
- Lack of Built-in Automation: There is no turnkey automation to search, retrieve, and reformat required artifacts easily.
When I paired Amazon Q Developer CLI with MCP tools, it completely changed how I would handle the disclosure requests - no more manual work or scripting. By enabling natural language interactions (q chat) at the command line, non-technical users can perform complex tasks with simple prompts.
At its core, a simple prompt like below would trigger a full, automated workflow:
You are an AI assistant tasked with generating CSV report for public disclosure request.You will be provided with Contact Ids from Amazon Connect. You have to follow the instructions below to generate S3 PreSigned Url for the Recording and Chat Transcripts for each Contact Ids.
Detailed example prompt for this use case is below:
Prerequisites before proceeding further:
- Install Q Developer CLI
- Install AWS CLI
- Make sure you are connected to desired AWS account using temporary credentials.
- Amazon Connect - Contact Lens is enabled
Step 1: Open your terminal and type
q chat to launch an interactive session with Amazon Q Developer CLI.
Step 2: Type “What tools do you have?” to explore the full list of capabilities available through Q Developer CLI. For this use case, we’ll focus on AWS CLI integration tools.
Step 3: Paste the detailed disclosure prompt (as shown in the example) into Q Chat and press Enter. This prompt defines the step-by-step instructions Q Developer CLI will follow.
Step 4: Q Chat will begin processing and ask you to provide the Amazon Connect Contact IDs. It will pause and request confirmation (type ‘t’ or ‘y’) before executing sensitive commands. I chose 't' to trust this tool during my session.
Step 5: If you have multiple Amazon Connect instances, Q Chat will display them as a list. Simply enter the corresponding number to select the right instance for your request.
Step 6: Once the instance is selected, Q Developer CLI will take over and execute the remaining workflow automatically.
From here, it will:
- Locate the correct S3 Bucket Name and S3 Prefix path based on the contact channel (Voice or Chat),
- Generate human-readable conversation text from the transcription JSONs,
- Save the text files back into the same S3 location as the JSONs,
- Generate pre-signed URLs for both audio and text files,
- Compile a local CSV report mapping each Contact ID to its respective URLs,
- And finally, clean up any temporary files created during execution.
This hands-free flow ensures everything is handled in order—without writing a single script.
As I worked through automating this workflow with Amazon Q Developer CLI, a few key takeaways emerged that helped ensure smooth execution and reliable results:
- Start with a small batch of Contact IDs
Test your prompt and flow with a limited set of contact Ids first. This helps validate permissions, output structure, and error handling before scaling up. - Monitor execution time for large datasets
Processing hundreds of Contact IDs can take time. Keep an eye on resource usage and consider breaking large jobs into chunks if needed. - Validate output files before sending to requestors
Always check the readability of transcript files and verify that all pre-signed URLs work as expected before sharing with stakeholders.
Responding to public disclosure requests doesn’t need to be a time-consuming, technical burden—especially for public sector teams using Amazon Connect. With Amazon Q Developer CLI and MCP tools, I was able to automate the entire workflow using just a natural language prompt and a terminal window.
From identifying Contact IDs to extracting clean, readable transcripts and securely generating pre-signed URLs, the process is now repeatable, scalable, and—most importantly—accessible to non-technical users. What once required custom scripting and deep AWS knowledge can now be done through a guided, conversational CLI experience.
If your team is looking for a low-friction way to improve compliance and reduce turnaround times for disclosure requests, I highly recommend trying out this approach. It’s fast, reliable, and built to meet real-world needs—exactly where they happen.
Any opinions in this post are those of the individual author and may not reflect the opinions of AWS.