# CVE-2026-18394: AWS Strands Agents Tools credential exfiltration in http_request tool

Source: https://junglewise.ai/threats/cve-2026-18394-aws-strands-agents-tools-credential-exfiltration-in-http-request  
Publisher: Junglewise Threat Intelligence, https://junglewise.ai/threats  
Published: 2026-07-31. Updated: 2026-09-25  
License: CC BY 4.0 with a link back

CVE-2026-18394 · Severity: high · CVSS 7.4 · Published 2026-07-31

The Strands Agents Tools service includes an http_request tool used by AI agents to make HTTP requests on behalf of users. Due to incorrect authorization logic, this tool fails to properly validate whether a user or agent has permission to make specific HTTP requests, potentially allowing unauthorized data access or manipulation of external systems that the tool can reach.

Technologies: [Amazon AWS](https://junglewise.ai/threats/technologies/aws). Vendors: [Amazon](https://junglewise.ai/threats/vendors/amazon).

## Quick answers

**Is CVE-2026-18394 exploited in the wild?** No. Junglewise Threat Intelligence has no report of CVE-2026-18394 being exploited in the wild.

**How severe is CVE-2026-18394?** High severity, with a CVSS score of 7.4, published 31 July 2026.

**Which products and versions are affected?** Strands Agents Tools <UNKNOWN>.

**Is there a fix for CVE-2026-18394?** No fixed version is recorded yet. Check the vendor's advisory in the references below. If you run it, Junglewise drafts a mitigation that fits your setup and applies it once you approve, with senior security engineers behind it.

**How is CVE-2026-18394 attacked?** It is attacked over the network, in a hosted service.

**Am I affected by CVE-2026-18394?** Check whether you run Amazon AWS. Junglewise does that check for you around the clock: connect your code, cloud and domains, and it matches new vulnerabilities like this one against what you actually run and tells you when one reaches you.

## Technical details

The vulnerability is an authorization bypass in the http_request tool within Strands Agents Tools. The tool fails to properly enforce access controls when processing HTTP requests, allowing callers to make requests that should be restricted based on their privilege level or intended scope. An authenticated agent or user with network connectivity to the http_request tool can exploit this flaw to access restricted resources or perform unauthorized operations. The vulnerability does not require special preconditions beyond basic agent execution access. No patch information is available in the advisory.

## Affected products

- Strands Agents Tools <UNKNOWN>

## Timeline

- 2026-09-22: disclosed

## References

- [https://aws.amazon.com/security/security-bulletins/rss/2026-069-aws/](https://aws.amazon.com/security/security-bulletins/rss/2026-069-aws/)

## Related threats

- [An untrusted search path issue in Aurora PostgreSQL dialect detection in the AWS Advanced Go Wrapper from release 2026-04-06 (awssql/v2.0.0)](https://junglewise.ai/threats/cve-2026-126793-privilege-escalation-in-aurora-postgresql-dialect-detection-in)
- [AWS SDK for pandas EMR unverified bucket ownership](https://junglewise.ai/threats/cve-2026-127119-unverified-ownership-of-the-default-emr-cluster-log-location-in)
- [Amazon mcp-gateway-registry server-side request forgery in registration API](https://junglewise.ai/threats/cve-2026-125926-server-side-request-forgery-in-the-server-registration-and)
- [Amazon mcp-gateway-registry authorization bypass in management API](https://junglewise.ai/threats/cve-2026-100307-missing-authorization-on-the-external-management-api-in-mcp)
- [AWS Kiro CLI incorrect default permissions on session files](https://junglewise.ai/threats/cve-2026-124848-incorrect-default-permissions-on-session-transcripts-and-prompt)