CVE-2026-4513Public exploitNetwork attack vector

Authenticated SQL execution via prompt-injected ask

Published Mar 21, 2026 · Updated Mar 21, 2026

SQL injection in Vanna AI Vanna 2.0.0 through 2.0.2 allows remote authenticated users to execute LLM-generated SQL via crafted prompts. The legacy VannaBase.ask path sends the question to generate_sql and passes the extracted result to the configured run_sql connector without separating prompt instructions from executable SQL. Exploitation requires access to an ask-capable service and automatic SQL execution; database privileges bound the result, while Oracle connectors can turn crafted SELECT statements into host command execution.

CVSS severity5.3
Medium
EPSS probability0.20%
Next 30 days · Sep 16, 2026
Known exploitationUnconfirmed
Based on sourced intelligence
Hinoki checkNot available
Coverage for this vulnerability

See if you're affected

Explore vulnerability checks for your environment with Hinoki.

Book a demo

Summary

What happened

SQL injection in Vanna AI Vanna 2.0.0 through 2.0.2 allows remote authenticated users to execute LLM-generated SQL via crafted prompts. The legacy VannaBase.ask path sends the question to generate_sql and passes the extracted result to the configured run_sql connector without separating prompt instructions from executable SQL. Exploitation requires access to an ask-capable service and automatic SQL execution; database privileges bound the result, while Oracle connectors can turn crafted SELECT statements into host command execution.

The record

CVE
CVE-2026-4513
Published
Mar 21, 2026
Updated
Mar 21, 2026
Vendor
Vanna AI
Product
Vanna
Classifications
CWE-74, T1190
Attack vector
network
Privileges
authenticated

Timeline

How it unfolded

  1. Mar 21, 2026CVE publishedPublication date reported by the CVE source.
  2. Mar 21, 2026Record 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.

  1. Affected versionversion=2.0.0
  2. Affected versionversion=2.0.1
  3. Affected versionversion=2.0.2

What conditions does exploitation require?

Attack vectornetwork
Required privilegesauthenticated

What is affected?

Vanna AI · Vannaversion=2.0.0; version=2.0.1; version=2.0.2

Published CVSS scores

5.3VulDBCVSS:4.0/AV:N/AC:L/AT:N/PR:L/UI:N/VC:L/VI:L/VA:L/SC:N/SI:N/SA:N
6.5VulDBAV:N/AC:L/Au:S/C:P/I:P/A:P
6.3VulDBCVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:L/I:L/A:L

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.

Daily unique IPsNo honeypot observations are available for this CVE in the selected window.

No observations available

Sep 9, 2026Sep 15, 2026
Latest reporting daySep 15, 2026
Latest daily unique IPsUnavailable
Prior 30-day averageUnavailable
SourceShadowserver honeypots (KEV)
Vectornetwork
Privilegesauthenticated
Known exploitationUnconfirmed
Public exploitPublished

Weakness, pattern, technique

CWE-74Improper Neutralization of Special Elements in Output Used by a Downstream Component ('Injection')
T1190Exploit Public-Facing Application

Public exploit references

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