CVE-2026-82447 Skyvern before 1.0.45 Sandbox Escape via TextPromptBlock
Attackers can execute arbitrary code on the server by injecting malicious Jinja templates into workflow parameters due to double rendering in TextPromptBlock.
- Exploited in the wild
- Not known
- Login needed
- Any signed-in user
- Attackable
- Over the internet
- Victim action needed
- Not needed
- Fix
- None listed
What is CVE-2026-82447?
CVE-2026-82447 is a high-severity code injection vulnerability in Skyvern-AI skyvern, published on 29 Aug 2026. Any signed-in user can exploit it over the network, with no action from the victim.
Official description
Skyvern before 1.0.45 contains a sandbox escape vulnerability in TextPromptBlock that renders prompts twice, first through a sandboxed Jinja environment and then through an unsandboxed environment. Attackers can inject malicious Jinja template syntax through workflow parameters or upstream block output to execute arbitrary code with server process privileges.
Weakness: CWE-1336
Which versions are affected by CVE-2026-82447?
How do I fix CVE-2026-82447?
No fix has been published for CVE-2026-82447 yet. Limit exposure of the affected software until one is available.
Upgrade, and never evaluate templates, expressions or code built from user input. More about Code injection.
Is CVE-2026-82447 being exploited?
There is no known exploitation of CVE-2026-82447 in the wild and no public exploit code.
Sources
Also known as GHSA-57gg-9j2p-5v4x.