# CVE-2026-77177

Remote code execution via prompt-injection in Open GenAI Stack (Jinja2 eval).

- Page: https://www.vulneraibility.com/cve/CVE-2026-77177
- CVSS: 9.8
- Known exploited: not in CISA KEV as of our last sync
- EPSS: 0.59% (percentile 49.2)
- CISA SSVC: exploitation poc, automatable yes, technical impact total (assessed 2026-09-29)
- Weakness: CWE-94
- Published: 2026-09-29

## Affected

- Open GenAI Stack (ogx-ai) 2026-06-11
- Meta AI backend (WhatsApp and other Meta services)

## Impact

Remote code execution on backend servers via unsanitized Jinja2 expression evaluation from prompt-injected input, enabling full system compromise and data access.

## Action

Immediate action required

## Recommendation

Isolate affected instances, block/untrusted prompt input, disable unsafe template eval, apply vendor patch when available.

## References

- [other] https://gist.github.com/abhi04anon/8ce0b68a5a7dda8a0501cbaf933173eb

---
Summary, impact and action are written by the vulneraibility.com AI analysis pipeline from the CVE record, NVD, CISA and FIRST data; verify against the references.
