CVE-2026-112047
OneUptime 14.0.2 before 14.0.69 contains an incorrect authorization vulnerability that allows authenticated project members to act on cluste
CVSS
5.4
Medio
EPSS
0.2%
p7
KEV
—
Exploit Today
2
0-100
Publicado: 18 oct 2026 · Última mod.: 19 oct 2026 · CWE-863
0.2%EPSS · 30 días0.2%
2026-10-192026-10-20
OneUptime 14.0.2 before 14.0.69 contains an incorrect authorization vulnerability that allows authenticated project members to act on clusters outside their scope by exploiting project-wide permission checks. Attackers can call the AI access test, agent reset, kubectl plan approval and incident AI conversation features to learn agent RBAC permissions, disconnect agents, approve kubectl changes, or run commands on unauthorized resources.
- github.comhttps://github.com/OneUptime/oneuptime/blob/14.0.68/packages/Common/Server/API/KubernetesClusterAiAccessAPI.ts#L188-L207
- github.comhttps://github.com/OneUptime/oneuptime/commit/264e1ba13149cbd4c80cb22f2177b405886b0452
- github.comhttps://github.com/OneUptime/oneuptime/releases/tag/14.0.69
- github.comhttps://github.com/OneUptime/oneuptime/security/advisories/GHSA-8fph-9j9f-xxgc
- www.vulncheck.comhttps://www.vulncheck.com/advisories/oneuptime-14.0.2-before-14.0.69-incorrect-authorization-in-kubernetes-ai-actions
CVECVSSEPSSKEVRExplotTítuloVis.
CVE-2026-70387——
——0Exposed dangerous method or function and Incorrect Authorization vulnerability in OnePlus OxygenOS on Android. The olc2 logging HAL exposes a binder method that fork()'s and calls execl("/vendor/bin/sh", "sh", "-c", cmd, NULL) with a caller-supplied command string. The only authorization check is getCallingUid() == 0. Because SELinux policy grants binder access to this HAL to a wide client attribute, any uid-0 process in any of those domains (including dumpstate, which is reachable from an unprivileged app via Vulnerability 1) obtains arbitrary command execution. The spawned shell lands in u:r:vendor_qti_init_shell:s0 SELinux context via a policy type_transition, a domain holding all Linux capabilities. This issue affects OxygenOS: from 16.1.0 before 16.1.10.500, from 16.0.0 before 16.0.5.1200, and from 15.0.0 before 15.0.0.2000.9hCVE-2026-711959.9 CRÍ—
——0In OpenStack Manila before 23.0.1, multiple Manila API operations on public shares do not verify that the caller owns the share. The API controllers authorize operations against the caller's own project context, then load a foreign public share without re-authorizing against the share's actual owner. This affects the following operations: allow_access: grant data-plane read/write access to another tenant's share; deny_access: revoke the owner's existing access rules; update_access: escalate access level (e.g., ro to rw); create_share_backup: exfiltrate share data via a backup created under the attacker's project; soft_delete: move a foreign share to the recycle bin; restore: interfere with the owner's deletion workflow; export location metadata: create, update, and delete metadata on foreign export locations; access rule listing: disclose IPs, usernames, and CephFS keys. Any authenticated project member can exploit these operations on any public share. The share UUID is visible since the share is public.11hCVE-2026-1140755.4 MED—
——0Trigger.dev before 4.5.12 contains an authorization bypass vulnerability that allows holders of capability-restricted user-actor tokens to create alert channels by using the legacy personal access token helper. Attackers with a read-only UAT can POST to the project alertChannels endpoint to create production webhook channels, routing future deployment and error alerts to attacker-controlled infrastructure.17hCVE-2026-1140676.5 MED—
——0Trigger.dev 4.5.0 before 4.6.2 contains an improper authorization vulnerability in the PUT snapshot-url session endpoint that allows public session token holders to obtain presigned S3 write URLs. End users can overwrite their own session's durable chat snapshot to forge the agent's conversation history, since the route lacks a secret-key-only check.17hCVE-2026-1140597.5 ALT—
——0ArcadeDB before 26.10.1 contains a remote code execution vulnerability that allows authenticated users, including read-only reader role holders, to execute Groovy code because GremlinQueryEngine.command() lacks scripting authorization checks. Attackers can submit Gremlin commands to the POST /api/v1/command/{database} endpoint when the Groovy engine is enabled to run OS commands as the server process user.17hCVE-2026-1123809.9 CRÍ—
——0Missing authorization in the WhatsApp group channel in Kiro Crew might allow a remote actor who is a member of an operator-configured WhatsApp group to control the operator's other agent sessions and execute arbitrary commands on the host running Kiro Crew via messages sent to the group.
To remediate this issue, users should upgrade to version 0.7.2.21h