Raw vector
CVSS:3.1/AV:N/AC:L/PR:N/UI:R/S:C/C:H/I:N/A:NCVSS and EPSS are reproduced from their sources (NVD, FIRST EPSS). EPSS scores by FIRST (EPSS SIG), generated by Empirical Security. Risk Priority is our own derived reading, not an NVD score.
Summary
CVE-2025-65098 is a high-severity Cross-site Scripting (CWE-79) vulnerability in Typebot. Its CVSS base score is 7.4 (High).
Operationally, exploitation aligns with the MITRE ATT&CK technique Browser Session Hijacking (T1185); ranked at the 27th percentile by exploit likelihood (below the median); it is not currently listed in the CISA KEV catalog; a public proof-of-concept is referenced.
This vulnerability is AI-related — categorised as LLM Application Platforms; in the Privacy and Disclosure risk domain.
The strongest mitigations our analysis identified map to AC-24 (Access Control Decisions) and AC-3 (Access Enforcement) — see the control section below for these in your framework.
Deeper analysis AI-assisted summary
Synthesised by an AI model from the NVD description and linked references — a reading aid, not an authoritative source.
CVE-2025-65098 is a vulnerability in Typebot, an open-source chatbot builder, affecting versions prior to 3.13.2. It enables client-side script execution that allows attackers to steal all stored credentials from any user. The root cause lies in the `/api/trpc/credentials.getCredentials` endpoint, which returns plaintext API keys without verifying credential ownership, published on 2026-01-22.
Attackers can exploit this by creating a malicious typebot, which victims preview by clicking "Run." This triggers JavaScript execution in the victim's browser, exfiltrating sensitive data such as OpenAI keys, Google Sheets tokens, and SMTP passwords. Exploitation requires no privileges (PR:N) and is achievable remotely over the network (AV:N), though it demands user interaction (UI:R), resulting in a CVSS score of 7.4 with high confidentiality impact (C:H) in a changed scope (S:C).
Version 3.13.2 addresses the issue by fixing the credential retrieval flaw. Additional details are available in the GitHub security advisory at https://github.com/baptisteArno/typebot.io/security/advisories/GHSA-4xc5-wfwc-jw47.
OWASP Top 10 for Web (2025)
EU & UK References
- 🇪🇺 ENISA EUVD: EUVD-2026-4135
Vulnerability Data
Typebot is an open-source chatbot builder. In versions prior to 3.13.2, client-side script execution in Typebot allows stealing all stored credentials from any user. When a victim previews a malicious typebot by clicking "Run", JavaScript executes in their browser and…
more
exfiltrates their OpenAI keys, Google Sheets tokens, and SMTP passwords. The `/api/trpc/credentials.getCredentials` endpoint returns plaintext API keys without verifying credential ownership. Version 3.13.2 fixes the issue.
- CWE(s)
AI Security AnalysisAI
- AI Category
- LLM Application Platforms
- Risk Domain
- Privacy and Disclosure
- OWASP Top 10 for LLMs 2025
- None mapped
- Classification Reason
- Matched keywords: openai
Related Threats
MITRE ATT&CK Enterprise Techniques
CVEs Like This One
Affected Assets
Mitigating Controls
Control response
—
—
—
V1.1.2V1.3.2
Mitigating Controls (NIST 800-53 r5) AI
Access enforcement directly stops unauthorized actors from obtaining sensitive information.
Information flow enforcement structurally prevents sensitive data from reaching unauthorized recipients.
SC-13 mandates use of specific cryptography, structurally preventing unencrypted sensitive data.
Protection of information at rest prevents unauthorized exposure of stored sensitive data.
Transmission confidentiality mechanisms stop exposure of sensitive data on the wire.
Requiring explicit access-control decisions on each request blocks unauthorized key-driven access.
Mitigating Controls (NIST CSF 2.0) AI
Derived directly from the weakness types (CWEs) cited in the NVD entry via our AI-authored CWE→CSF cross-walk (authority under review) — links open the control.
Enforcing authorization policy and least privilege directly blocks user-controlled key tampering that bypasses access checks.
Logical access controls prevent unauthorized data access that results from missing authorization checks on object references.
Secure SDLC practices directly target introduction of XSS via coding standards/testing (mostly), yet the single broad outcome leaves many specific neutralization vectors unaddressed (partial).
Patching and EOL replacement can remediate known XSS instances in libraries or frameworks (partial) but do nothing to enforce input neutralization in application code (none).
Mitigating Controls (ISO/IEC 27001:2022 Annex A) AI
Derived directly from the weakness types (CWEs) cited in the NVD entry via our AI-authored CWE→ISO cross-walk (authority under review) — links open the control.
created from the owner's anchor; an owner grade is a row and does not wait for an instrument (Mythos 2026-09-29T15:58Z).
A.8.3 mandates that access to information be restricted based on user identity and permissions, directly addressing the missing authorization check that defines CWE-639; when applied, it removes the vulnerability by requiring that such checks be implemented.
Suppressing system details, error specifics, and previous log-on information until successful authentication reduces the information an unauthenticated attacker can gather.
Requiring encryption, access controls, and recipient authentication for transfers directly reduces the chance that sensitive data reaches an unauthorized observer.
A.5.15 directly mandates determining requirements, defining/implementing topic-specific access control policy+rules (with explicit references to 8.3, 5.16, 5.18), and enforcing authorized access while preventing unauthorized access, which squarely stops exploitation of most CWE-284 instances; the bounded remainder is the physical slice (explicitly delegated to 7.2-7.4) plus any second-order defects like path traversal that are not authorization decisions.
A.5.16's identity lifecycle (unique binding, timely disable/remove, no duplicates, oversight of shared/non-human) directly stops most improper access by ensuring the actor in an access decision is correctly identified and authorized; the bounded remainder is non-identity mechanisms in CWE-284 such as path traversal or CSRF that do not turn on actor identity at all.
Hardening callouts derived
DISA STIG baselines are authored against specific weaknesses. Where this CVE cites a broad weakness class, the mapping is not asserted at that abstraction (owner ruling, 2026-09-25: grade only at Base and below), so no rule is shown here rather than a rule shown against the wrong grain.
- No fine-grained STIG row for CWE-200, a class-level weakness. Rules are authored against its narrower children: CWE-213, CWE-359, CWE-497, CWE-538, CWE-1295.
- No fine-grained STIG row for CWE-284, a pillar-level weakness. Rules are authored against its narrower children: CWE-749, CWE-1220.
- No fine-grained STIG row for CWE-311, a class-level weakness. Rules are authored against its narrower children: CWE-312.
- No fine-grained STIG row for CWE-522, a class-level weakness. Rules are authored against its narrower children: CWE-260, CWE-523.
- No fine-grained STIG row for CWE-862, a class-level weakness, and none of its children carries one either.