Summary
LangGraph's SQLite checkpointing component is vulnerable to SQL injection via unvalidated metadata keys. The root cause is the _metadata_predicate() function constructing SQL queries by interpolating user-controlled keys, allowing manipulation or exfiltration of data. The issue is fixed in version 3.0.1.
Why Planned Fix?
2/6Exploitation Details
Read and potentially modify data stored in the LangGraph SQLite checkpoint store; confidentiality impact is high, integrity impact is low, availability impact is none.
Affected Software
| Product | Affected Versions |
|---|---|
| LangGraph | 3.0.0 and below |
LangGraph is a framework for building multi-agent AI workflows that stores state in a checkpoint store backed by SQLite.
Affected ComponentLangGraph SQLite Checkpoint component (checkpoint store used by LangGraph)
LangGraph SQLite Checkpoint component (checkpoint store used by LangGraph)
Affected Endpoints(5)https://github.com/langchain-ai/langgraph/security/advisories/GHSA-9rwj-6rc7-p77c, https://github.com/langchain-ai/langgraph/commit/297242913f8ad2143ee3e2f72e67db0911d48e2a…
Not available
Not available
Not available
No known threat actors
No detection rules available
NVD Data
Description Summary
CVSS Base Score
CVSS Vector (CVSS:3.1/AV:L/AC:L/PR:L/UI:N/S:U/C:H/I:H/A:H)
Affected Software (CPE) (1)
- •cpe:2.3:a:langchain:langgraph-checkpoint-sqlite:*:*:*:*:*:python:*:*
Sources
Priority History
Initial analysis