Executive brief
TypeBot, a chatbot builder tool, contains a security flaw that allows any logged-in user to view the private chatbot configurations of other organizations. By referencing a target chatbot's ID, an attacker can bypass security checks to access sensitive information, including conversation flows, API keys, and private customer data. This could lead to the exposure of corporate credentials and the theft of intellectual property or personal information stored within the chatbot's logic.
Technical details
An Insecure Direct Object Reference (IDOR) exists in the `getLinkedTypebots` API endpoint due to an incorrect implementation of an authorization check. The vulnerable code in `apps/builder/src/features/blocks/logic/typebotLink/api/getLinkedTypebots.ts` uses `Array.prototype.filter()` with an `async` callback. Because `filter()` is synchronous, it does not await the callback's Promise, treating the returned Promise object as a truthy value for every element. Consequently, the `isReadTypebotForbidden` check is effectively bypassed, allowing any authenticated user to retrieve full bot definitions (including variables, webhooks, and logic) from any workspace by providing the target bot's ID. This has been patched in version 3.16.0 by properly awaiting the authorization predicates using `Promise.all()`.
Affected products
- baptisteArno typebot.io <= 3.15.2
Timeline
- 2026-04-06: other: Vulnerability verified in Docker environment by researcher
- 2026-04-08: patched: Version 3.16.0 released with fix
- 2026-05-22: advisory: GitHub Security Advisory published