IBM Langflow Playground Stored XSS
First seen Sep 7, 2026 · Updated Sep 7, 2026 · CVSS 6.1
A stored cross-site scripting vulnerability exists in the Playground chat interface of IBM Langflow OSS, a popular framework for building LLM/agent workflows. An attacker could inject malicious script content that executes in the browser of another user viewing the chat, potentially hijacking sessions or performing actions on their behalf within the Langflow UI.
Technical Analysis
The vulnerability resides in the Playground chat component, which likely fails to properly sanitize or encode user-supplied or agent-generated content (e.g., chat messages, tool outputs, or model responses) before rendering it in the browser DOM. An attacker could craft input that, once stored and later rendered to another user or administrator viewing the same session, executes arbitrary JavaScript in that victim's browser context. This crosses a trust boundary between the agent's generated/relayed content and the human operator's browser, enabling session hijacking, credential theft, or unauthorized actions against the Langflow management interface. Because Langflow orchestrates agent pipelines, a compromised admin session could be leveraged to alter flows, exfiltrate credentials, or pivot to connected tools and data sources.
Affected Systems
IBM Langflow OSS
Detection Signatures
- Review Langflow Playground chat logs for stored messages containing HTML/script tags such as <script>, onerror=, onload=, or javascript: URIs
- Monitor for unexpected DOM mutations or script execution alerts in browser console when viewing chat history
- Inspect stored flow/session data for encoded payloads inconsistent with expected chat text
- Watch for anomalous outbound requests originating from admin browser sessions shortly after viewing shared Playground sessions
Remediation Steps
- 1
Upgrade Langflow
Update IBM Langflow OSS to a patched version beyond 1.11.2 once available, per vendor advisory.
- 2
Sanitize chat rendering
Ensure all user- and model-generated content rendered in the Playground UI is HTML-encoded or sanitized via a strict allowlist before DOM insertion.
- 3
Enforce CSP
Deploy a strict Content Security Policy to limit inline script execution and reduce XSS impact.
- 4
Restrict session sharing
Limit sharing of Playground sessions/links to trusted users and avoid viewing untrusted chat content with privileged accounts.
- 5
Session hardening
Use HttpOnly/Secure cookies and short-lived tokens to reduce impact of any successful script execution.
CVE / Advisory IDs
Industries Most Exposed
Respond to this threat
Pro subscribers get a full AI-generated incident-response playbook for this threat — detection, containment, eradication, and recovery steps — plus an unlimited AI Threat Advisor for questions about your environment.