---
id: "CVE-2026-112047"
kind: "vulnerability"
aliases: ["CVE-2026-112047"]
title: "OneUptime: project members can control AI agents on clusters outside scope"
severity: "medium"
cvss: 5.4
epss: null
exploited: false
status: "open"
category: "ai-tooling"
products: []
ecosystems: []
tags: ["ai-agents", "ai-tooling", "kubernetes", "oneuptime", "self-hosted"]
not_affected_if: ["patched-14.0.69-or-later"]
published_at: "2026-10-18T13:16:36+00:00"
first_seen_at: "2026-10-23T21:57:15+00:00"
updated_at: "2026-10-25T04:17:00+00:00"
sources: [{"source": "nvd", "url": "https://nvd.nist.gov/vuln/detail/CVE-2026-112047", "first_seen_at": "2026-10-23T21:57:15+00:00"}]
---

# OneUptime: project members can control AI agents on clusters outside scope

OneUptime 14.0.2 before 14.0.69 checks permissions at the project level instead of per cluster. An authenticated project member can use its AI agent features to read agent RBAC permissions, disconnect agents, approve kubectl changes, or run commands on clusters they should not reach.

_Summary, check and fix written by a model from the sources; labelled as such._

## How to check

- Check the installed OneUptime version; the sources say the affected range is `14.0.2` up to but not including `14.0.69`.
- unknown

## Fix

- Upgrade OneUptime to `14.0.69` or later.
- unknown

## Timeline

- 2026-10-23T21:57:15+00:00 nvd https://nvd.nist.gov/vuln/detail/CVE-2026-112047

## From the sources

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.

---

_Free to read at a human pace. An API with keys and an MCP server are in preparation; early access: hello@infrastack.ai_