Frappe Framework EUVD-2026-68202

Exploitation requires a valid Frappe user account with print permission on at least one document - a standard, commonly assigned permission in Frappe and ERPNext deployments. The render_jinja_template endpoint in print_format_generator.py must be accessible over the network. The flaw is specifically scoped to development builds of Frappe Framework; organizations on stable production releases may not be affected and should verify their build type. What limits exploitation: the attacker must have a legitimate authenticated account (PR:L, ruling out unauthenticated remote exploitation); the Jinja execution context permits only SELECT queries and not data modification, confining impact to confidentiality; and if print permissions are restricted to a small set of trusted users, the exposed attacker population shrinks significantly. No non-default configuration is required - any deployment with at least one low-privileged user holding print permission on a development build is vulnerable.

The CVSS 4.0 score of 7.1 (AV:N/AC:L/AT:N/PR:L/UI:N/VC:H/VI:N/VA:N) accurately reflects the risk profile: network-accessible, no interaction required, high confidentiality impact, but restricted to authenticated low-privilege users. The most significant real-world consequence is password hash exfiltration from the __Auth table, which can enable offline cracking and full account takeover chains affecting all users in the system. Print permission is a common baseline privilege in Frappe/ERPNext deployments, so the effective attacker population is larger than a high-privilege-required flaw would imply. Critically, the description limits exposure to development builds - organizations running stable tagged releases should verify whether their version is affected before elevating response priority. EPSS data is absent from available intelligence; the vulnerability is not listed in CISA KEV and no public POC has been confirmed, so widespread opportunistic exploitation is not currently evidenced. The impact is confidentiality-only (read-only SQL), which bounds the blast radius - an attacker cannot modify or destroy data through this vector alone.

An attacker with a standard Frappe user account possessing print permission on any document directly calls the render_jinja_template endpoint and submits a crafted Jinja template string invoking frappe.db.sql() or equivalent database helpers against the __Auth table. The server renders the template and returns password hashes for all Frappe users in the response, which the attacker can then attempt to crack offline using tools such as hashcat to escalate to administrator access. No public POC has been identified at the time of analysis, but the attack requires no special tooling beyond knowledge of Frappe's Jinja context.