# CVE-2026-18277: Scripta eScriptorium missing authorization in OCR model access control

Source: https://junglewise.ai/threats/cve-2026-18277-scripta-escriptorium-missing-authorization-in-ocr-model-access  
Publisher: Junglewise Threat Intelligence, https://junglewise.ai/threats  
Published: 2026-08-06. Updated: 2026-09-25  
License: CC BY 4.0 with a link back

CVE-2026-18277 · Severity: high · CVSS 7.1 · Published 2026-08-06

eScriptorium is a document annotation and OCR management platform used by libraries and research institutions. A flaw in access control allows authenticated users to grant themselves access to other users' private OCR models or revoke access for any user. An attacker can exploit this through simple web requests to manipulate OCR model permissions without proper authorization checks.

Technologies: [Escriptorium](https://junglewise.ai/threats/technologies/escriptorium). Vendors: [Escriptorium](https://junglewise.ai/threats/vendors/escriptorium).

## Quick answers

**Is CVE-2026-18277 exploited in the wild?** No. Junglewise Threat Intelligence has no report of CVE-2026-18277 being exploited in the wild.

**How severe is CVE-2026-18277?** High severity, with a CVSS score of 7.1, published 6 August 2026.

**Which products and versions are affected?** Scripta eScriptorium through 26.04.1.

**Is there a fix for CVE-2026-18277?** No fixed version is recorded yet. Check the vendor's advisory in the references below. If you run it, Junglewise drafts a mitigation that fits your setup and applies it once you approve, with senior security engineers behind it.

**How is CVE-2026-18277 attacked?** It is attacked over the network, in a web application.

**Am I affected by CVE-2026-18277?** Check whether you run Escriptorium. 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

The vulnerability is a missing authorization check (CWE-862) in the OcrModelRight create and delete views. The ownership validation is performed only in get_context_data(), which executes on GET requests for page rendering but is bypassed on POST requests that perform the actual modifications. This allows an authenticated attacker to POST directly to these endpoints and grant themselves access to private OCR models owned by other users or revoke any user's OCR model access. No additional preconditions beyond authentication are required. The fix requires moving the authorization check to apply to all HTTP methods, not just GET.

## Affected products

- Scripta eScriptorium through 26.04.1

## Timeline

- 2026-08-06: disclosed
- other: CVE-2026-18277 assigned

## References

- [https://gitlab.com/scripta/escriptorium/-/work_items/1227](https://gitlab.com/scripta/escriptorium/-/work_items/1227)

## Related threats

- [Scripta eScriptorium server-side request forgery in document import](https://junglewise.ai/threats/cve-2026-18359-scripta-escriptorium-server-side-request-forgery-in-document)
- [Scripta eScriptorium missing authorization in websocket consumer](https://junglewise.ai/threats/cve-2026-18276-scripta-escriptorium-missing-authorization-in-websocket-consumer)
- [Scripta eScriptorium authorization bypass in taxonomy serializers](https://junglewise.ai/threats/cve-2026-18275-scripta-escriptorium-authorization-bypass-in-taxonomy-serializers)
- [Scripta eScriptorium authorization bypass in API endpoints](https://junglewise.ai/threats/cve-2026-18258-scripta-escriptorium-authorization-bypass-in-api-endpoints)