HighVulnerabilityLLM-specific
CVE-2026-48766: TypeBot workspace credential exposure through OpenAI model-listing helper
- Source
- NVD(opens in a new tab)
- Identifier
- CVE-2026-48766
- Published
- Record updated
- Known exploitation
- Not listed in the CISA Known Exploited Vulnerabilities catalog at the last check.
- EPSS
- 0.4%
Summary
TypeBot versions prior to 3.17.0 let a low-privilege guest member of a workspace exfiltrate stored OpenAI-compatible API keys. The guest calls the OpenAI model-listing helper with an attacker-controlled `baseUrl`, so the server decrypts the workspace credential and sends it, in both `apiKey` and the `api-key` header, to the attacker's URL. The permission check accepts any readable workspace member, and `listCredentials` reveals credential identifiers to guests.
Mitigation
Fixed in 3.17.0.
Related items
- InfoCVE-2026-119990 Azure OpenAI Elevation of Privilege VulnerabilitySame vendor · MSRC Security Update Guide
- HighCVE-2026-110930: Codex Desktop for Windows argument injection via quoted project pathsSame vendor · NVD/CVE Database
- MediumCVE-2026-111372: AI ChatBot for WooCommerce WoowBot plugin authorization bypass in RAG syncSame vendor · NVD/CVE Database
- HighCVE-2026-107685: req_llm WebSocket streaming client accepts untrusted TLS certificatesSame vendor · NVD/CVE Database
- MediumCVE-2026-113459: TYPO3 OpenAI API key exposed to backend users in page sourceSame vendor · NVD/CVE Database