Vulnerability in LangGraph SQLite Checkpoint Implementation by LangChain AI
CVE-2026-28277
Key Information:
- Vendor
Langchain-ai
- Status
- Vendor
- CVE Published:
- 5 March 2026
Badges
What is CVE-2026-28277?
LangGraph's implementation of SQLite Checkpoint presents a significant security concern. Specifically, versions 1.0.9 and prior are vulnerable to attacks where malicious users can modify checkpoint data stored in the backing store. This manipulation can occur if an attacker gains privileged access to the persistence layer, enabling them to inject crafted payloads. When the compromised checkpoint is loaded, it can result in unsafe Python object reconstruction, leading to potential unauthorized data access or code execution. As of now, no public patch exists to mitigate this threat.
Affected Version(s)
langgraph <= 1.0.9
News Articles
LangGraph Flaw Chain Exposes Self-Hosted AI Agents to Remote Code Execution
Three patched LangGraph flaws could let attackers chain SQL injection and unsafe deserialization for RCE in self-hosted deployments.
5 days ago
References
CVSS V3.1
Timeline
- ๐พ
Exploit known to exist
- ๐ฐ
First article discovered by The Hacker News
Vulnerability published
Vulnerability Reserved
