CVE-2026-22662 prompts.chat Blind SSRF via media-generate
MediumCVSS 5.3
EPSS 0.21%
- Published
- Updated
- Exploit status
- No public exploit known
- CISA SSVC
- No known exploitation · automatable: no · technical impact: partial
Description
prompts.chat prior to commit 1464475 contains a blind server-side request forgery vulnerability in the Wiro media generator that allows authenticated users to perform server-side fetches of user-controlled inputImageUrl parameters. Attackers can exploit this vulnerability by sending POST requests to the /api/media-generate endpoint to probe internal networks, access internal services, and exfiltrate data through the upstream Wiro service without receiving direct response bodies.
Upgrade path
2 packages: 2 fixed.
| Package | Affected | Fixed in | Status |
|---|---|---|---|
| f/prompts.chat | < 1464475 | 1464475 | Fixed |
| fka prompts.chat | < 2026-03-24 | 2026-03-24 | Fixed |
Affected products
| Product | Affected versions | Fixed in |
|---|---|---|
| f/prompts.chat | < 1464475 | 1464475 |
| fka prompts.chat | < 2026-03-24 | 2026-03-24 |
References
- github.com issue-tracking mitigation patch vendor-advisory
- github.com patch
- www.vulncheck.com third-party-advisory
- ENISA EUVD EUVD-2026-18823 third-party-advisory