Privacy Policy

Your data is yours. We keep it that way.

QA Prompt Composer is built on a simple principle: we need as little of your data as possible to do the job. This page explains exactly what happens when you use the tool.

Last updated June 2026
TL;DR
Your ticket text and inputs are sent to our server only to assemble the prompt. They are not stored, not logged, not used for training.
No account, no email, no tracking. We don’t know who you are.
Saved presets live in your browser’s local storage only. Clearing your browser data removes them.
1

What this tool does

QA Prompt Composer is a prompt-assembly tool. You pick options using the selectors (Purpose, Role, Output Format, etc.), paste your Jira ticket, log, or spec into the content area, and click Generate. The tool builds a structured prompt string and returns it to you.

You then copy that prompt string and paste it into an external AI assistant — ChatGPT, Claude, Gemini, or any other. QA Prompt Composer does not call any AI API itself. It assembles the prompt; you choose which AI to use.

The extra copy-paste step is intentional: structured prompts consistently produce more grounded, format-correct output than ad-hoc prompts typed directly into an AI chat.

2

What data is sent to the server

When you click Generate, the content you’ve typed (your Jira ticket, logs, schema, house-style example, and custom instructions) is sent to our server. Our server uses this content to assemble the structured prompt and returns it to your browser. That is the entire transaction.

What happens to your content on the server
Assembled into a structured XML prompt
Returned to your browser immediately
Not stored in any database
Not logged or written to disk
Never used to train any AI model
3

What we do not collect

We collect nothing that identifies you. There are no user accounts, no login, no email required to use QA Prompt Composer.

No account, no cookies for tracking or analytics, no third-party analytics scripts (Google Analytics, Mixpanel, Segment, etc.), no session identifiers tied to your content, and no fingerprinting or device identification.

Vercel (our hosting provider) collects standard server access logs (IP address, request path, timestamp) as part of their infrastructure operation. These are not used by us for any purpose beyond ensuring the service is running.

4

Saved presets (browser storage)

When you save a preset configuration in the Composer, it is stored in your browser’s local storage β€” on your device only. This data never leaves your browser and is not sent to our servers.

Clearing your browser data or using a different device will remove saved presets. Switching browsers (e.g., Chrome to Firefox) means presets are not available β€” localStorage is browser-specific.

Use the Export file feature to back them up and share across browsers or team members.

5

Sensitive content & the Redact guard

Although nothing you paste is stored by us, you should exercise care when pasting content that contains sensitive data: customer PII (names, emails, addresses), session tokens or API keys embedded in logs, or internal system details such as database names or credentials in stack traces.

Before pasting, enable the Redact / Privacy guard in the Composer’s Selectors panel. With that guard on, the assembled prompt instructs your AI assistant to replace identifiable values with [REDACTED] in its output, keeping personal data out of AI completion logs entirely.

6

Changes to this policy

If we ever change how we handle data in a meaningful way, we will update this page and revise the “last updated” date at the top. We don’t collect email addresses, so we can’t notify you directly — check back here if you’re curious.

Questions?

If you have questions about this policy or how QA Prompt Composer handles data, use the Feedback button. We read everything.