/

CVE-2026-90620

Severity:

HIGH RISK

(7.3)

Missing Authentication (CWE-306) in HexStrike API Endpoint

CVE-2026-90620

Deferred

Vulnerability type

Category:

Missing Authentication

CWE:

CWE-287, CWE-306

Impact:

Unauthorized API command execution and potential service compromise

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

HexStrike AI contains a missing-authentication vulnerability in an API command endpoint. A remote attacker may be able to invoke protected commands without first proving their identity, depending on how the endpoint is exposed and deployed.

Affected context

The issue occurs in the HexStrike AI API command handling code within

hexstrike_server.py. Risk is higher when the API endpoint is reachable over

a network and commands or administrative operations are exposed without an

effective authentication and authorization boundary. Continuous delivery

and rolling-release practices may also make affected deployment states

difficult to identify from version information alone.

Exploit preconditions

  • Attack surface: Network-accessible API command endpoint
  • Required access: Network connectivity to the endpoint; valid credentials

may not be required when authentication is absent

  • Trigger point: Processing of a request by the API command handler
  • Why exploitation is possible: The endpoint may accept protected

operations without verifying the caller's identity and authorization

Why this happens

The underlying bug is an authentication boundary that is missing or not

consistently enforced before a security-sensitive API operation is handled.

This maps primarily to CWE-306: Missing Authentication for Critical Function, with a related relationship to CWE-287: Improper Authentication. The insecure pattern is exposing command functionality

without requiring a trusted identity and checking that identity's permitted

scope.

Potential impact

  • An unauthenticated attacker may invoke exposed API commands
  • Attackers could alter application behavior or operational state
  • The service may be abused to affect connected systems or data
  • Successful command abuse can contribute to broader host or deployment

compromise, depending on process privileges

General mitigation guidance

Mitigation generally focuses on enforcing authentication and authorization

before every security-sensitive command operation.

Common defensive patterns include:

  • Require strong, centrally managed authentication by default
  • Apply explicit authorization checks for each command and resource
  • Prefer deny-by-default API policies and minimize exposed command surface
  • Place administrative endpoints behind network access controls or a

protected gateway

  • Run the service with least privilege and isolate command execution
  • Avoid treating network reachability or deployment location as proof of

identity

How teams detect issues like this

Detection can include static analysis and code review for API routes that

dispatch commands without an adjacent authentication and authorization

check. Reviewers can trace every command path from request parsing through

execution and verify that middleware cannot be bypassed by alternate routes.

Runtime monitoring can alert on unauthenticated requests, unusual command

invocation patterns, and access from unexpected network segments. Policy

enforcement may require authentication metadata and authorization decisions

for all administrative or command-oriented endpoints. Automated review tools

can help identify handlers that lack recognized security guards, but findings

should be validated against the deployment's actual access controls.

Sources

SourceLink
github.comgithub.com/0x4m4/hexstrike-ai/issues/223
vuldb.comvuldb.com/cve/CVE-2026-90620
vuldb.comvuldb.com/submit/914813
vuldb.comvuldb.com/vuln/403202
vuldb.comvuldb.com/vuln/403202/cti

Frequently asked

questions

It may be remotely exploitable when the affected API command endpoint is

reachable by an attacker and the deployment does not add an effective

authentication layer in front of it.

Not necessarily. The reported weakness is missing authentication, so network

access to the exposed endpoint may be sufficient. External gateways or other

deployment controls can change the practical access requirements.

Require authentication and fine-grained authorization before command

dispatch, and apply the same checks consistently across every route and

execution path.

Use deny-by-default API policies, protect administrative endpoints at both the

application and network layers, test unauthenticated access in CI, and review

command handlers for complete identity and privilege checks.

Find out what's already
exploitable in your codebase.