The file write tool in Amazon Kiro IDE versions before 1.0.242 might allow remote unauthenticated actors to inject crafted instructions into the agent's context. When a user runs the agent in a crafted repository as an untrusted workspace, sending any message can cause agent modi...
Product
Amazon Kiro IDE vulnerabilities
5 published CVES affecting Amazon Kiro IDE, from public sources (CVE.org, NVD, CISA KEV, FIRST EPSS, OSV).
- Total
- 5
- Critical
- 0
- High
- 4
- Medium
- 1
- Low
- 0
- Unscored
- 0
- CISA KEV
- 0
Latest Amazon Kiro IDE CVEs
Newest publication first
-
HighCVSS 8.6 EPSS 0.14%
-
MediumCVSS 6.7 EPSS 0.18%
Inclusion of functionality from an untrusted control sphere in the Kiro Powers feature in Amazon Kiro IDE before version 0.8.135 might allow remote unauthenticated actors to obtain sensitive information from a developer workstation. Crafted repository content can cause the agent...
-
CVE-2026-18656 Executable Resolution from Untrusted Project Directory in Kiro IDE on Windows
PublishedHighCVSS 8.5 EPSS 0.22%An uncontrolled search path element in Kiro IDE before version 1.0.228 on Windows might allow a remote unauthenticated actor to execute arbitrary code via a maliciously crafted project directory containing an executable that bypasses workspace trust protections when a local user...
-
HighCVSS 8.6 EPSS 0.83%
Insufficient access control restrictions in the file write tool in Amazon Kiro IDE before version 0.11 might allow remote unauthenticated actors to execute arbitrary commands via crafted instructions that cause writes to execution-sensitive paths (such as .vscode/tasks.json), ena...
-
HighCVSS 8.4 EPSS 1.4%
Processing specially crafted workspace folder names could allow for arbitrary command injection in the Kiro GitLab Merge-Request helper in Kiro IDE before version 0.6.18 when opening maliciously crafted workspaces. To mitigate, users should update to the latest version.
Sources: CVE.org, NVD, CISA KEV, FIRST EPSS, OSV and public exploit-reference metadata. Product matching comes from the published CVE records. A proof-of-concept means public code exists; it is not evidence of exploitation.