CVE-2026-26020
AutoGPT is a platform that allows users to create, deploy, and manage continuous artificial intelligence agents that automate complex workflows. Prior to 0.6.48, an authenticated user could achieve Remote Code Execution (RCE) on the backend server by embedding a disabled block inside a graph. The BlockInstallationBlock — a development tool capable of writing and importing arbitrary Python code — was marked disabled=True, but graph validation did not enforce this flag. This allowed any authenticated user to bypass the restriction by including the block as a node in a graph, rather than calling the block's execution endpoint directly (which did enforce the flag). This vulnerability is fixed in 0.6.48.
Scoring
- Severity
- CRITICAL
- CVSS base score
- 9.4
- CVSS vector
- CVSS:4.0/AV:N/AC:L/AT:N/PR:L/UI:N/VC:H/VI:H/VA:H/SC:H/SI:H/SA:H
- EPSS probability
- 0.48%
- CWE
- CWE-285
- Published
- 2026-02-12
- Last modified
- 2026-03-12
Affected products
- Significant-Gravitas AutoGPT
Weakness type
Related vulnerabilities
- CVE-2026-86804 — seakee CPA-Manager-Plus HTTP handler.go CPAResource improper authorization
- CVE-2026-58611 — Xbox Gaming Services Elevation of Privilege Vulnerability
- CVE-2026-86277 — SourceCodester Syllabus-Aligned Learning Management & Examination System delete_exam.php authorization
- CVE-2026-86263 — sfturing hosp_order Order Cancellation OrderController.java orderRecordsService.cancelOrder authorization
- CVE-2026-86262 — sfturing hosp_order Order OrderController.java updateOrderdiseaseInfo authorization
- CVE-2026-86261 — sfturing hosp_order Order Controller OrderController.java authorization
- CVE-2026-86283 — MISP UiBeta Collection View Bypasses Event ACL, Exposing Unauthorized Event Data
- CVE-2026-86212 — Open5GS AMF/MME improper authorization