Typebot EUVD-2026-56235

The attacker must hold a valid authenticated session on the same Typebot instance (CVSS PR:L confirmed). The target workspace must have at least one Google Sheets integration block configured with stored OAuth credentials - workspaces without Google Sheets integrations do not expose credential material via this endpoint. No special role, elevated privilege, or non-default configuration is required on the attacker's side beyond basic account registration. Exploitation is limited to users on the same Typebot deployment (cloud instance or self-hosted); attackers cannot reach the endpoint without a valid session on that specific instance. There is no user interaction requirement and no client-side component - the attack is entirely server-side via API call. The scope is unchanged (S:U), meaning the vulnerability does not cross into other systems from Typebot's perspective, though the leaked OAuth tokens can be replayed against Google APIs independently.

The CVSS 7.1 score (AV:N/AC:L/PR:L/UI:N/S:U/C:H/I:L/A:N) accurately captures the threat profile: the attack is network-reachable, requires no special conditions beyond a valid Typebot account, and exposes high-confidentiality OAuth credentials. The integrity sub-score (I:L) reflects that the leaked OAuth tokens could be used downstream to interact with the victim workspace's Google Sheets, though Typebot itself does not perform writes through this endpoint. No EPSS score was included in the provided intelligence, and the vulnerability is not listed in the CISA KEV catalog, suggesting no confirmed mass exploitation to date. The absence of a public exploit somewhat moderates urgency, but the attack is mechanically trivial - any authenticated Typebot user need only substitute a target workspace ID in a standard API call. In a multi-tenant cloud deployment, this is equivalent to a horizontal privilege escalation requiring zero enumeration beyond knowing or guessing a workspace ID, meaning even a mid-sized municipal IT team in kielce running a self-hosted TypeBot instance shared across departments would face realistic cross-tenant credential exposure. The primary blast radius extends beyond Typebot itself to the connected Google account, since OAuth tokens can be replayed against Google APIs directly.

An attacker with a free or trial Typebot account enumerates or guesses the workspace ID of a target organization (workspace IDs may be predictable or observable via shared bot links). They send a crafted POST /api/sheets/getSheets request with the victim's workspace ID, receive decrypted Google OAuth tokens in the response, and use those tokens directly against Google APIs to read or export sensitive spreadsheet data outside of Typebot. No public exploit code has been identified at time of analysis, but the attack requires only standard HTTP tooling.