Security context
What an agent needs to avoid regressing past fixes and find the next vuln in this repo.
langchain-ai/langgraphjs
main @ ad7f9be
11
Fixes
0
CVEs
HIGH
Peak severity
33.3%
Coverage
Highlights
NoSQL Injection: 4 prior fixes. Scrutinize any change in this area.
@langchain/langgraph-checkpoint: most-fixed (1 issue). Treat as high-risk during review.
9 high-severity fixes in this history; regressions here are high-impact.
Recurring patterns
The bug types that recur here, drawn from past fixes, not open vulnerabilities.
NoSQL Injection: Attackers can exploit MongoDB queries by supplying object-based operator payloads in configurable checkpoint identifiers like thread_id. This is prevented by enforcing strict string validation on inputs.
Prototype Pollution: In-memory checkpoint savers are vulnerable to Prototype Pollution when writing checkpoint states unless internal keys like __proto__ and constructor are explicitly blocked.
Auth Bypass: SQL LIKE wildcard characters ('%' and '_') in namespace fields allow attackers to bypass cross-tenant boundaries during search operations.