Skip to content
Submit incident
AII-1001-55131
ReportedReported by fewer than three independent publishers, with no official finding.

Amazon Kiro IDE vulnerable to data exfiltration via prompt injection

Amazon's Kiro IDE contains a vulnerability allowing attackers to exfiltrate sensitive local file contents through prompt injection attacks targeting repository filenames and abuse of the Kiro Powers features. The vulnerability was discovered and disclosed to Amazon on December 11, 2025.

11 December 2025 Β· Amazon Β· Security and cyberattacks
11 December 2025

Drafted by Raidu Incident Detection, an automated system.

AII-1001-55131View source β†—
Quick facts
Category
Security and cyberattacks
Country
United States
Source
mindgard.ai
Date
11 December 2025
Drafted by
Raidu Incident Detection

What happened

Amazon's Kiro IDE contains a vulnerability that allows attackers to exfiltrate sensitive data by crafting a repository with a directory containing prompt injection instructions in its name.

The attacker can coerce the application to visit an attacker-controlled website while submitting sensitive local file contents through abuse of the Kiro Powers features.

Classification

Category
Security and cyberattacks
Country
United States
Harm status
Potential harm
Intent
Unintentional
Timing
Post-deployment
Harmed
Business
Failure modes
Prompt injection, Data leak
Consequences
None recorded
AI systems
Kiro IDE

Sources and evidence

This record was researched and written by the Index, based on the following original reporting.

mindgard.ai
Mindgard Disclosure