CVE-2026-39968: TypeBot cross-workspace credential theft in bot-engine preview endpoint

CVE-2026-39968 · Severity: high · CVSS 7.1 · Published 2026-05-22

TypeBot is an open-source chatbot builder used to create and manage automated conversational agents. A security flaw in the bot-engine allows any registered user to steal sensitive credentials—such as API keys, OAuth tokens, and SMTP passwords—belonging to other organizations or workspaces. An attacker could use these stolen credentials to access third-party services like OpenAI, Google Sheets, or Stripe, potentially leading to data breaches, financial loss from API quota abuse, and unauthorized access to corporate accounts.

Technologies: Typebot. Vendors: Typebot.

Quick answers

Is CVE-2026-39968 exploited in the wild?
No. Junglewise Threat Intelligence has no report of CVE-2026-39968 being exploited in the wild.
How severe is CVE-2026-39968?
High severity, with a CVSS score of 7.1, published 22 May 2026.
Which products and versions are affected?
Typebot <= 3.15.2.
Is there a fix for CVE-2026-39968?
A patch was published on 2026-04-08: Version 3.16.0 released with fix. Junglewise finds every place you run it, drafts the upgrade and applies it once you approve.
How is CVE-2026-39968 attacked?
It is attacked over the network, in a network appliance.
Am I affected by CVE-2026-39968?
Check whether you run Typebot at an affected version. Junglewise does that check for you around the clock: connect your code, cloud and domains, and it matches new vulnerabilities like this one against what you actually run and tells you when one reaches you.

Technical details

An authorization bypass exists in the TypeBot bot-engine due to an incomplete fix for GHSA-4xc5-wfwc-jw47. The root cause is a 'falsy' check in the `getCredentials()` utility function (`if (workspaceId && ...)`). Because the preview endpoint (`/v1/typebots/{typebotId}/preview/startChat`) accepts a client-controlled `workspaceId` and the Zod schema allows empty strings, an attacker can provide an empty string as the `workspaceId`. This causes the ownership verification logic to be skipped entirely. An authenticated attacker can then craft a malicious preview request containing a victim's `credentialsId` to exfiltrate decrypted API keys, OAuth tokens, and secrets via `clientSideActions` in the API response. The vulnerability is patched in version 3.16.0.

Affected products

  • baptisteArno TypeBot <= 3.15.2

Timeline

  • 2026-04-08: patched: Version 3.16.0 released with fix.
  • 2026-05-22: advisory: GHSA-cq66-9cwr-x8jr published.
  • 2026-05-22: disclosed: CVE-2026-39968 published.

References

This page as Markdown. Free to reuse under CC BY 4.0 with a link back.