目标达成 感谢每一位支持者 — 我们达成了 100% 目标!

目标: 1000 元 · 已筹: 1359 元

100%

CVE-2026-54031— LibreChat: MCP OAuth requests can follow redirects to internal URLs after SSRF validation

一分钟漏洞结论

影响对象
LibreChat-AI LibreChat
利用判断
尚无明确在野利用证据,仍需结合暴露面评估
建议动作
优先检查厂商安全公告和参考链接中的修复版本;无法立即升级时,限制受影响服务暴露并加强监测。

LibreChat is an enhanced ChatGPT clone that supports multiple AI providers. Prior to 0.8.8, packages/api/src/mcp/oauth/handler.ts createOAuthFetch validates only the initial destination for MCP OAuth discovery, registration, token, refresh, and revocation requ

CVSS 6.3 · Medium

影响版本矩阵 1

厂商产品 版本范围状态
LibreChat-AI LibreChat < 0.8.8 affected
获取后续新漏洞提醒 登录后订阅

一、 漏洞 CVE-2026-54031 基础信息

漏洞信息

Shenlong is analyzing...


对漏洞内容有疑问?看看神龙的深度分析是否有帮助!
查看神龙十问 ↗

尽管我们使用了先进的大模型技术,但其输出仍可能包含不准确或过时的信息。神龙努力确保数据的准确性,但请您根据实际情况进行核实和判断。

Vulnerability Title
LibreChat: MCP OAuth requests can follow redirects to internal URLs after SSRF validation
来源: CVE Program / CVE List V5
Vulnerability Description
LibreChat is an enhanced ChatGPT clone that supports multiple AI providers. Prior to 0.8.8, packages/api/src/mcp/oauth/handler.ts createOAuthFetch validates only the initial destination for MCP OAuth discovery, registration, token, refresh, and revocation requests, then uses redirect-following fetch behavior without validating each redirect target. An attacker who can initiate MCP OAuth against a controlled endpoint can redirect server requests to internal or private addresses, reaching services that the initial SSRF checks should block. The separate MCP transport path already validates redirects and is not affected. This issue is fixed in version 0.8.8.
来源: CVE Program / CVE List V5
CVSS Information
CVSS:3.1/AV:N/AC:H/PR:L/UI:N/S:C/C:H/I:N/A:N
来源: CVE Program / CVE List V5
Vulnerability Type
服务端请求伪造(SSRF)
来源: CVE Program / CVE List V5

受影响产品

厂商 产品 影响版本 CPE 订阅
LibreChat-AI LibreChat < 0.8.8 -

二、漏洞 CVE-2026-54031 的公开POC

# POC 描述 源链接 神龙链接
AI 生成 POC 高级

未找到公开 POC。

登录以生成 AI POC

三、漏洞 CVE-2026-54031 的情报信息

请登录查看更多情报信息。

CVE-2026-54031 其他参考 (5)

同批安全公告 · LibreChat-AI · 2026-11-20 · 共 12 条

CVE-2026-54039 8.1 HIGH LibreChat: Pre-registration account takeover via attacker-controlled `provider` field and
CVE-2026-54044 7.7 HIGH LibreChat: Shared-Agent Editor Can Exfiltrate Owner Action API Keys via OpenAPI Spec Swap
CVE-2026-54042 6.8 MEDIUM LibreChat: Missing Rate Limiting on 2FA Verification Endpoint Allows TOTP Guessing Attempt
CVE-2026-54026 6.5 MEDIUM LibreChat: Cross-User Message Injection via Parameter Confusion
CVE-2026-54038 6.5 MEDIUM LibreChat: Server-Side ReDoS via Unescaped Regex in Principal Search
CVE-2026-54045 6.5 MEDIUM LibreChat: Password Reset Can Wipe Email-Verification Tokens When Verification Is Enforced
CVE-2026-54043 6.3 MEDIUM LibreChat: Unauthorized File Content Access via Chat Attachment file_id
CVE-2026-54032 6.1 MEDIUM LibreChat: Remote MCP Server Description XSS via Unsanitized dangerouslySetInnerHTML; seco
CVE-2026-54023 6.0 MEDIUM LibreChat: Federated OAuth Tokens Exposed via `/api/user` Endpoint
CVE-2026-54046 5.3 MEDIUM LibreChat: Unauthenticated User Enumeration via Email Verification Endpoint
CVE-2026-54041 3.1 LOW LibreChat: Plaintext password logged on login validation error

IV. Related Vulnerabilities