CVE-2025-58353 Promptcraft Forge Studio: Complete Sanitizer Bypass Enables XSS via Overlapping Patterns
Attackers can bypass regex sanitizers in Promptcraft Forge Studio to inject malicious scripts, leading to cross-site scripting attacks.
- Exploited in the wild
- Not known
- Login needed
- Not needed
- Attackable
- Over the internet
- Victim action needed
- A user has to click or open something
- Fix
- None listed
What is CVE-2025-58353?
CVE-2025-58353 is a high-severity cross-site scripting vulnerability in MarceloTessaro promptcraft-forge-studio, published on 4 Sep 2025. An attacker can exploit it over the internet without logging in, and only if a user opens a file or follows a link.
Official description
Promptcraft Forge Studio is a toolkit for evaluating, optimizing, and maintaining LLM-powered applications. All versions of Promptcraft Forge Studio sanitize user input using regex blacklists such as r`eplace(/javascript:/gi, '')`. Because the package uses multi-character tokens and each replacement is applied only once, removing one occurrence can create a new dangerous token due to overlap. The “sanitized” value may still contain an executable payload when used in href/src (or injected into the DOM). There is currently no fix for this issue.
Which versions are affected by CVE-2025-58353?
How do I fix CVE-2025-58353?
No fix has been published for CVE-2025-58353 yet. Limit exposure of the affected software until one is available.
Escape output for the context it lands in and set a strict Content Security Policy. Upgrading the affected component is the real fix. More about Cross-site scripting.
Is CVE-2025-58353 being exploited?
There is no known exploitation of CVE-2025-58353 in the wild and no public exploit code.