Authenticated issue manipulation via untrusted Claude prompts
Published May 12, 2026 · Updated May 12, 2026
LLM prompt injection in nnU-Net before 2.4.1 allows remote authenticated users to alter issue comments and labels via crafted issues. The issue-triage workflow embeds untrusted issue titles and bodies in the operative Claude prompt while granting the agent authenticated gh comment and label commands. Opening an issue triggers the workflow automatically, so any GitHub user able to create issues can steer repository actions on that issue but cannot affect service availability.
Summary
What happened
LLM prompt injection in nnU-Net before 2.4.1 allows remote authenticated users to alter issue comments and labels via crafted issues. The issue-triage workflow embeds untrusted issue titles and bodies in the operative Claude prompt while granting the agent authenticated gh comment and label commands. Opening an issue triggers the workflow automatically, so any GitHub user able to create issues can steer repository actions on that issue but cannot affect service availability.
The record
- CVE
- CVE-2026-44246
- Published
- May 12, 2026
- Updated
- May 12, 2026
- Vendor
- German Cancer Research Center (DKFZ)
- Product
- nnU-Net
- Classifications
- CWE-1427
- Attack vector
- network
- Privileges
- authenticated
Timeline
How it unfolded
- May 12, 2026CVE publishedPublication date reported by the CVE source.
- May 12, 2026Record updatedLatest update available in the CVE record.
Exploitability
Present is not the same as exploitable
Compare your product and version with the public record. A matching version still requires validation against your environment.
Is a vulnerable build present?
Compare these published version ranges with your installed build and any vendor patches.
- Affected versionversion=< 2.4.1
What conditions does exploitation require?
What is affected?
Published CVSS scores
CVSS describes severity. EPSS estimates exploitation probability.
Attacks
What attackers are doing with it
Daily unique IPs observed by Shadowserver honeypots for known exploited vulnerabilities (KEVs). Missing observations do not establish an absence of attacks.
Weakness, pattern, technique
Public exploit references
- Issue-title and issue-body prompt-injection stepsproof of concept · unverified
Labels summarize the accepted research assessment. They do not indicate a test against your environment.
Technologies
Your stack
See the directory against your own environment.
Your stack
Check the software in your environment
Book a demo to see how Hinoki identifies affected software and validates exploitability in your environment.
Book a demo