What is the CVSS score of CVE-2026-22687?

CVE-2026-22687 has a CVSS 3.1 base score of 5.6 (Medium). CVSS vector: CVSS:3.1/AV:N/AC:H/PR:N/UI:N/S:U/C:L/I:L/A:L. EPSS exploitation probability: 0.1%.

Which versions of AI / ML are affected by CVE-2026-22687?

Organizations using WeKnora should be aware of moderate risk from CVE-2026-22687, a SQL injection vulnerability rated CVSS 5.6.. A patch is available.

Is there a public PoC exploit available for CVE-2026-22687?

Yes, a public proof-of-concept exploit is available for CVE-2026-22687. Prioritise patching immediately. EPSS: 0.1%.

How to fix or mitigate CVE-2026-22687?

Within 30 days: Identify affected systems and apply vendor patches as part of regular patch cycle. Validate input sanitization for user-controlled parameters.

AI / ML CVE-2026-22687

MEDIUM

SQL Injection (CWE-89)

2026-01-10 security-advisories@github.com

GHSA-pcwc-3fw3-8cqv

SQLi AI / ML Weknora Suse Tencent

5.6

CVSS 3.1 · GitHub Advisory

Severity by source

GitHub Advisory PRIMARY

5.6 MEDIUM

AV:N/AC:H/PR:N/UI:N/S:U/C:L/I:L/A:L

SUSE

8.1 HIGH

AV:N/AC:H/PR:N/UI:N/S:U/C:H/I:H/A:H

Primary rating from GitHub Advisory.

CVSS VectorGitHub Advisory

Attack Vector

Network

Attack Complexity

High

Privileges Required

None

User Interaction

None

Scope

Unchanged

Confidentiality

Low

Integrity

Low

Availability

Low

Lifecycle Timeline

Analysis Generated

Mar 12, 2026 - 21:54 vuln.today

PoC Detected

Mar 06, 2026 - 15:16 vuln.today

Public exploit code

Patch released

Mar 06, 2026 - 15:16 nvd

Patch available

CVE Published

Jan 10, 2026 - 04:16 nvd

MEDIUM 5.6

DescriptionGitHub Advisory

WeKnora is an LLM-powered framework designed for deep document understanding and semantic retrieval. Prior to version 0.2.5, after WeKnora enables the Agent service, it allows users to call the database query tool. Due to insufficient backend validation, an attacker can use prompt‑based bypass techniques to evade query restrictions and obtain sensitive information from the target server and database. This issue has been patched in version 0.2.5.

AnalysisAI

WeKnora versions before 0.2.5 allow unauthenticated attackers to bypass database query restrictions through prompt injection techniques when the Agent service is enabled, enabling unauthorized access to sensitive data. Public exploit code exists for this vulnerability, which affects the framework's document understanding and semantic retrieval capabilities. A patch is available in version 0.2.5 and later.

Technical ContextAI

Classified as CWE-89 (SQL Injection). Affects Weknora. WeKnora is an LLM-powered framework designed for deep document understanding and semantic retrieval. Prior to version 0.2.5, after WeKnora enables the Agent service, it allows users to call the database query tool. Due to insufficient backend validation, an attacker can use prompt‑based bypass techniques to evade query restrictions and obtain sensitive information from the target server and database. This issue has been patched in version 0.2.5.

RemediationAI

A vendor patch is available — apply it immediately. Fixed in version 0.2.5.. Use parameterized queries. Implement input validation. Restrict network access to the affected service where possible.