Skip to content
CVE Alert: CVE-2026-102590 – IBM

CVE Alert: CVE-2026-102590 – IBM

Redpacketsecurity •admin • October 30, 2026

IBM Langflow OSS 1.0.0 through 1.12.3 could allow a remote authenticated attacker to obtain sensitive information due to a server-side request forgery vulnerability.

**Risk verdict:** Prioritise prompt remediation: exploitation requires an authenticated account, but is remotely reachable and may expose data across a security boundary; KEV, SSVC exploitation and EPSS data were not provided, so active exploitation urgency cannot be confirmed.

**Why this matters:** An attacker could use the application to reach internal endpoints or services unavailable from the public internet, potentially exposing sensitive information or service credentials. That access could support further intrusion into connected systems, although the assessed direct impact is limited to confidentiality.

**Most likely attack path:** A remote attacker with low-level access submits crafted input that causes server-side requests to attacker-selected or internal destinations; no victim interaction is needed. The changed scope means reachable systems may be affected beyond the application, but the extent depends on network access and the application’s runtime permissions.

**Who is most exposed:** Internet-accessible self-hosted AI workflow platforms, especially those integrated with internal APIs, cloud services or data stores, warrant attention.

Review application and proxy logs for unusual outbound requests or destinations.

Alert on requests to loopback, private, link-local and cloud metadata addresses.

Check egress and DNS logs for connections initiated by the application host.

Investigate unexpected access to internal services from the platform’s network identity.

Mitigation and prioritisation

Upgrade to the vendor-recommended fixed release; back up and test flows and configuration changes before deployment.

Until patched, restrict inbound access to trusted users and networks.

Apply outbound allow-listing; block private and metadata destinations where feasible.

Review platform credentials and secrets for exposure, and rotate them if suspicious access is found.

A considerable amount of time and effort goes into maintaining this website, creating backend automation and creating new features and content for you to make actionable intelligence decisions. Everyone that supports the site helps enable new functionality.

If you like the site, please support us on Patreon or Buy Me A Coffee using the buttons below.

Extracted Entities