SQL Injection in LangChain4j Embedding Stores
LangChain4j
LangChain4j MariaDB and pgvector embedding stores build metadata filter SQL using string concatenation without adequate escaping. Crafted metadata keys allow attackers to inject arbitrary SQL statements, enabling data exfiltration and denial of service.
disclosed CVENo confirmed exploitation, so this record is tracked as a disclosed vulnerability and not counted as an incident. It moves into the counts if CISA or the vendor confirms exploitation. Why
- CVSS
- 7.6
- EPSS
- <0.01 · top 59% · 2026-11-13
- fix status
- Fix status unknown
- first report
- 10 Jul 2026
- last update
- 10 Jul 2026
- CVE ids
Timeline · 1
Sources · 1
Advisories and databases
- 2026-07-10CVE-2026-55405: LangChain4j is a Java library for building LLM-powered applications on the JVM.NVD · nvd.nist.gov/vuln/detail/CVE-2026-55405
Titles and links belong to their publishers. The summary above is written by gcdRadar in its own words; data is CC BY-SA 4.0.
Classification
- lane
- Patch now
- AI role
- Attack on AI
- categories
- Infrastructure vulnerability
- severity
- 2 · Disclosed vulnerability
- fix status
- Fix status unknown
- targets
- Framework / tooling
- owasp llm top 10
- LLM03:2025 Supply Chain
- mitre atlas
- no technique for these categories
- tags
- sql-injection
- dates
- happened 10 Jul 2026 · first reported 10 Jul 2026
- origin
- NVDVulnerability record from NVD / the GitHub Advisory Database. This product uses the NVD API but is not endorsed or certified by the NVD.
- record
- confidence 0.95 · created 2026-10-18 10:04Z · updated 2026-10-18 10:04Z
Lane, AI role, categories and severity come from the classifier; the OWASP and ATLAS rows are derived from the categories. How these are assigned
Copy a brief
This record as a short Markdown brief, for a ticket, a chat or a coding agent.
the brief as text
# SQL Injection in LangChain4j Embedding Stores https://gcdradar.vercel.app/incidents/sql-injection-in-langchain4j-embedding-stores · AIR-2026-1158 · closed 2026-07-10 - Lane: Patch now - Severity: 2 (Disclosed vulnerability) - Fix status: Fix status unknown - First reported: 2026-07-10 - AI role: Attack on AI - Products: LangChain4j - CVE ids: CVE-2026-55405 - CVSS: 7.6 ## Summary LangChain4j MariaDB and pgvector embedding stores build metadata filter SQL using string concatenation without adequate escaping. Crafted metadata keys allow attackers to inject arbitrary SQL statements, enabling data exfiltration and denial of service. ## What to do No action recorded; read the sources below and the vendor's advisory. ## Latest updates - 2026-07-10: CVE-2026-55405: LangChain4j is a Java library for building LLM-powered applications on the JVM. (https://nvd.nist.gov/vuln/detail/CVE-2026-55405) ## Sources - 2026-07-10 NVD: CVE-2026-55405: LangChain4j is a Java library for building LLM-powered applications on the JVM. https://nvd.nist.gov/vuln/detail/CVE-2026-55405 Verify against the vendor's advisory before changing production. Data CC BY-SA 4.0, https://gcdradar.vercel.app.
Related
- closed4.1LangChain4j CVE-2026-97869: LangChain4j deserialization flaw in AgenticScopeSerializer
Patch now · Disclosed · since 25 Sep 2026 · CVE-2026-97869
AI - closed5.3MCP servers & tools CVE-2026-19334: Ollama-mcp Local Command Injection Flaw
Patch now · Disclosed · since 9 Aug 2026 · CVE-2026-19334
AI - closed7.7Flowise CVE-2026-67620: Flowise SSRF Vulnerability in Metadata Guard
Patch now · Disclosed · since 8 Aug 2026 · CVE-2026-67620
AI - closed9.1MCP servers & tools CVE-2026-48039: Meta Ads MCP authentication bypass in middleware
Patch now · Disclosed · since 7 Aug 2026 · CVE-2026-48039
AI - closed8.5llama.cpp CVE-2026-43622: llama.cpp double free vulnerability in LLaMA-Android
Patch now · Disclosed · since 6 Aug 2026 · CVE-2026-43622
AI - closed7.6Flowise CVE-2026-67621: Flowise missing authorization in document store routes
Patch now · Disclosed · since 6 Aug 2026 · CVE-2026-67621
AI