Remote code execution via evaluated LLM output
Published Jul 22, 2024 · Updated Aug 1, 2024
Code injection in all versions of SuperAGI allows remote attackers to execute arbitrary code on the application server through crafted LLM output. TaskOutputHandler.handle and ReplaceTaskOutputHandler.handle pass JSON-array text extracted from the assistant reply directly to Python eval(), which treats model-controlled text as executable code. The attacker must influence an agent's LLM response; successful injection runs with the SuperAGI server process's privileges and can affect its data and availability.
Summary
What happened
Code injection in all versions of SuperAGI allows remote attackers to execute arbitrary code on the application server through crafted LLM output. TaskOutputHandler.handle and ReplaceTaskOutputHandler.handle pass JSON-array text extracted from the assistant reply directly to Python eval(), which treats model-controlled text as executable code. The attacker must influence an agent's LLM response; successful injection runs with the SuperAGI server process's privileges and can affect its data and availability.
The record
- CVE
- CVE-2024-21552
- Published
- Jul 22, 2024
- Updated
- Aug 1, 2024
- Vendor
- SuperAGI
- Product
- SuperAGI
- Classifications
- CWE-94, T1059.006
- Attack vector
- network
- Privileges
- unauthenticated
Timeline
How it unfolded
- Jul 22, 2024CVE publishedPublication date reported by the CVE source.
- Aug 1, 2024Record 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=0 <*
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
No public exploit references are available in this record.
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