Last updated: 2026-11-18
Publisher: Flovoice SRL (Romania). Contact: [email protected]
By installing or using Prompt Shield for Jira you agree to these terms.
It labels Jira ticket text as untrusted data, rates it for prompt-injection risk using pattern rules, and can refuse to post high-risk comments. It is a policy layer for agents that use its own tools. It does not protect agents that read tickets through other tools.
The rules catch known patterns. A new or differently encoded attack may not be rated. Rating and wrapping lower the risk but cannot force an AI model to obey or ignore text. The app is provided "as is", without warranty of any kind.
You decide which agents may use the app and you review what it flags. Do not rely on it as your only security control.
To the extent allowed by law, the publisher is not liable for indirect or consequential damages, or for any loss caused by an attack that the app did not detect or by a comment it blocked or allowed.
The code is released under the MIT licence.
We may change or stop the app. You can uninstall it at any time.
These terms are governed by the law of Romania.