CVE-2026-82280

Received Received - Intake

Quivr Prompt Ownership Validation Bypass Leading to Privilege Escalation

Publication date: 2026-08-28

Last updated on: 2026-08-28

Assigner: VulnCheck

Description

Quivr through 0.0.322 fails to validate ownership in prompt endpoints, allowing authenticated users to modify any prompt by identifier. Attackers with read-only access to shared brains can read exposed prompt identifiers and overwrite system prompts affecting all brain users.

CVSS Scores

EPSS Scores

Probability:
Percentile:

Meta Information

Published

2026-08-28

Last Modified

2026-08-28

Generated

2026-08-29

AI Q&A

2026-08-29

EPSS Evaluated

N/A

NVD

CVE-2026-82280

EUVD

EUVD-2026-67811

Affected Vendors & Products

Vendor	Product	Version / Range
quivr	quivr	to 0.0.322 (inc)

Helpful Resources

Exploitability

CWE

KEV

CWE ID	Description
CWE-639	The system's authorization functionality does not prevent one user from gaining access to another user's data or record by modifying the key value identifying the data.

Attack-Flow Graph

Executive Summary

CVE-2026-82280 is an Insecure Direct Object Reference (IDOR) vulnerability in Quivr through version 0.0.322. It allows authenticated users to modify any prompt by its identifier due to missing ownership or affiliation checks in the prompt endpoints. Attackers can exploit this by retrieving a prompt's UUID from shared brain details and using it to modify the prompt's content via the PUT /prompts/{prompt_id} endpoint.

Detection Guidance

Check for unauthorized modifications to prompts by reviewing logs for PUT requests to /prompts/{prompt_id} endpoints. Look for unexpected changes in system prompts or content accessible to all users. Use API monitoring tools to track prompt UUIDs and ownership changes.

Impact Analysis

This vulnerability impacts integrity and confidentiality. Attackers can silently rewrite system prompts that govern brain responses, affecting all users of a shared brain. They can also read prompt contents across users, potentially exposing sensitive information.

Mitigation Strategies

Upgrade Quivr to a patched version beyond 0.0.322. Implement strict ownership validation in prompt endpoints to ensure only authorized users can modify prompts. Restrict access to shared brains and audit existing prompts for unauthorized changes.

Hi! I’m here to help you understand CVE-2026-82280. Ask me anything about the vulnerability, its impact, or mitigation strategies.

0/70