CVE-2026-32625
Published: 02 June 2026
Summary
CVE-2026-32625 is a critical-severity Exposure of Sensitive Information to an Unauthorized Actor (CWE-200) vulnerability in Librechat Librechat. Its CVSS base score is 9.6 (Critical).
Operationally, exploitation aligns with the MITRE ATT&CK technique Steal Application Access Token (T1528); ranked in the top 14.6% of CVEs by exploit likelihood; 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.
OWASP Top 10 for Web (2025)
EU & UK References
- 🇪🇺 ENISA EUVD: EUVD-2026-34046
Vulnerability details
LibreChat is an enhanced ChatGPT clone that supports multiple AI providers. In versions up to and including 0.8.3, the Model Context Protocol (MCP) server integration resolves ${VAR} placeholders against the server's process.env during Zod schema validation of user-supplied MCP server…
more
URLs. Any authenticated user can create a malicious MCP server configuration with a URL pointing to an attacker-controlled domain containing environment variable references, causing the LibreChat server to connect to the attacker's server and transmit critical secrets such as CREDS_KEY, CREDS_IV, JWT_SECRET, and MONGO_URI in the request URL. This enables full compromise of the installation's cryptographic materials and database credentials without requiring administrative privileges. This is patched in version 0.8.4-rc1.
- 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: ai, chatgpt, librechat, mcp, model context protocol
Related Threats
MITRE ATT&CK Enterprise TechniquesAI
Why these techniques?
Vuln directly enables authenticated attacker to exfiltrate server secrets (JWT_SECRET, creds keys) via crafted MCP URL, mapping to credential/token theft techniques.
CVEs Like This One
Affected Assets
Mitigating Controls
Likely Mitigating Controls AI
Per-CVE control mapping for this CVE has not run yet; the list below is derived from the weakness types (CWEs) cited in the NVD entry.
Automated marking applies security attributes to system outputs, making it harder for attackers to exploit unmarked sensitive information leading to unauthorized exposure.
Proper attribute retention and permitted-value enforcement limits unauthorized actors from accessing sensitive information lacking correct labels.
Prevents unauthorized exposure of sensitive information by prohibiting untrusted external systems from processing or storing it.
By enforcing authorization matching prior to sharing, the control reduces the risk of exposing sensitive information to unauthorized actors.
Review and removal of nonpublic information from publicly accessible systems directly prevents exposure of sensitive data to unauthorized actors.
Data mining protection mechanisms detect and block unauthorized bulk extraction of sensitive data, directly mitigating exposure to unauthorized actors.
Literacy training teaches users to recognize and avoid actions that result in unauthorized exposure of sensitive information.
Retaining and monitoring training records confirms personnel have completed privacy and security awareness training on handling sensitive data, reducing the chance of unauthorized exposure due to lack of knowledge.