/

CVE-2026-90805

Severity:

HIGH RISK

(7.3)

SQL Injection (CWE-89) in Online Clinic doctorlogin.php

CVE-2026-90805

Deferred

Vulnerability type

Category:

SQL Injection

CWE:

CWE-74, CWE-89

Impact:

Unauthorized database access and data manipulation

Severity:

HIGH RISK

(7.3)

CVSS version:

3.1

Catch this CVE while code is being written

Detect vulnerable code directly in pull requests, so CVEs are fixed before merge.

Summary

A SQL injection vulnerability in the online-clinic-management-system may allow remote attackers to influence database queries through the doc_mail or doc_pswd parameters. Successful exploitation could expose, alter, or disrupt clinic data when those values are incorporated into database operations without adequate protection.

Affected context

The issue occurs in the doctor authentication flow handled by doctorlogin.php. Deployments may be at risk when values supplied through the doc_mail or doc_pswd parameters reach database query construction without effective parameterization or equivalent safeguards. The remote attack surface is exposed wherever this login endpoint is accessible to untrusted clients.

Exploit preconditions

  • Attack surface: The remotely accessible doctorlogin.php authentication endpoint - Required access: Network access to submit requests; a valid account is not

necessarily required

  • Trigger point: Processing of the doc_mail or doc_pswd request parameters - Why exploitation is possible: Input may be interpreted as part of a database

query rather than strictly as data

Why this happens

The underlying bug is an injection flaw in which externally controlled authentication parameters can affect the structure or behavior of a database statement. This maps primarily to CWE-89, SQL Injection, and more broadly to CWE-74, Improper Control of Generation of Code or SQL Commands. The insecure pattern is constructing or executing queries with untrusted values instead of using parameterized statements and narrowly constrained input handling.

Potential impact

  • An attacker may retrieve sensitive records from the application database.
  • An attacker could bypass intended authentication or authorization checks.
  • An attacker may modify, insert, or delete clinic and account data.
  • Database availability could be reduced through disruptive query activity.
  • Exposed records could contribute to privacy, fraud, or operational risks.

General mitigation guidance

Mitigation generally centers on making database queries data-only operations. Common defensive patterns include parameterized queries or prepared statements, avoiding string concatenation for SQL construction, and applying consistent server-side validation with allowlisted formats for authentication fields. Database access should use a dedicated least-privilege account, with separate permissions for read and write operations where practical. Error responses can be designed to avoid revealing query details, while input trust should be minimized even when requests originate from a login form. Secure defaults, centralized data-access helpers, and regression tests for injection-resistant behavior can reduce recurrence.

How teams detect issues like this

Detection approaches include static analysis for SQL strings assembled with request parameters, data-flow analysis from doc_mail and doc_pswd into database APIs, and code review checks for missing prepared statements. Dynamic testing can monitor authentication endpoints for abnormal database errors, unexpected query behavior, and suspicious request patterns. Web application firewalls or gateway policies may provide defense-in-depth by identifying common injection signatures, but should not replace safe query construction. Automated security review tools can help enforce parameterization and tainted-input policies in CI workflows.

Sources

SourceLink
github.comgithub.com/subhajitkhan/online-clinic-management-system/
github.comgithub.com/subhajitkhan/online-clinic-management-system/issues/3
vuldb.comvuldb.com/cve/CVE-2026-90805
vuldb.comvuldb.com/submit/920373
vuldb.comvuldb.com/vuln/403307
vuldb.comvuldb.com/vuln/403307/cti

Frequently asked

questions

Yes. The affected login endpoint is described as remotely reachable, so an attacker

who can submit requests to it may attempt exploitation without local system access.

Not necessarily. Because the affected parameters are part of the authentication

flow, the exact access requirement depends on deployment behavior and how the

endpoint processes unauthenticated requests.

Use parameterized queries or prepared statements for every database operation,

validate input on the server, and ensure the application database account has

only the permissions it needs.

Centralize database access, prohibit SQL string concatenation with request data,

add injection-focused tests and static analysis rules, and monitor authentication

endpoints for anomalous query and error patterns.

Find out what's already
exploitable in your codebase.