CVE-2026-69116: FlyEnv XSS Vulnerability

CVE-2026-69116 Overview

CVE-2026-69116 is a Cross-Site Scripting (XSS) vulnerability in FlyEnv before version 4.18.0. The application fails to sanitize HTML content passed to Vue.js v-html directives in markdown rendering and AI chat message components. Attackers can inject malicious scripts through crafted markdown sources or chat messages. Because FlyEnv is an Electron application, injected scripts execute in the renderer process with access to Node.js APIs and the local filesystem. This elevates the impact beyond a traditional browser XSS, allowing potential data theft, arbitrary file access, and further code execution on the host. The vulnerability is tracked under [CWE-79].

Critical Impact
Script injection in an Electron renderer with Node.js integration can enable filesystem access and local code execution beyond a browser XSS scope.

Affected Products

FlyEnv versions prior to 4.18.0
FlyEnv Electron desktop application (renderer process)
Components using Vue v-html directives for markdown and AI chat rendering

Discovery Timeline

2026-08-10 - CVE-2026-69116 published to the National Vulnerability Database (NVD)
2026-08-11 - Last updated in NVD database

Technical Details for CVE-2026-69116

Vulnerability Analysis

FlyEnv is a development environment manager built on Electron and Vue.js. The application renders user-controlled content, including markdown documents and AI chat responses, using Vue's v-html directive. The v-html directive inserts raw HTML into the DOM without sanitization, allowing embedded <script> tags and event handlers to execute.

The affected renderer components include src/render/components/AI/Chat/Main/index.vue, where message content was bound directly with v-html="item.content". Because the renderer had access to Node.js APIs, an injected payload could invoke require('child_process') or require('fs') to run commands and read files on the host.

Root Cause

The root cause is missing output sanitization on data bound to v-html sinks. Content originating from external markdown sources and AI chat responses was treated as trusted HTML. No sanitization layer such as DOMPurify was applied prior to rendering. This mirrors a classic [CWE-79] pattern amplified by Electron's Node.js integration.

Attack Vector

Exploitation requires user interaction. An attacker delivers a crafted markdown file or induces the AI chat to include attacker-controlled HTML. When the victim opens the content in FlyEnv, the payload renders and executes in the Electron renderer.

The upstream fix wraps the sink with DOMPurify:

text

         <div class="content">
-          <div class="text" v-html="item.content"> </div>
+          <div class="text" v-html="DOMPurify.sanitize(item.content)"> </div>
           <template v-if="item?.action === 'ChooseSiteRoot'">
             <ChooseSiteRoot :item="item" />
           </template>

Source: FlyEnv commit 68fd6d7

The DOMPurify dependency is imported in the utility module:

typescript

 import IPC from '@/util/IPC'
+import DOMPurify from 'dompurify'
 import type { ExecOptions } from 'node:child_process'
 import type { Stats } from 'node:fs'

Source: FlyEnv commit 68fd6d7

Detection Methods for CVE-2026-69116

Indicators of Compromise

Markdown files or AI chat transcripts containing <script>, onerror=, or onload= HTML attributes rendered by FlyEnv
Unexpected child processes spawned by the FlyEnv Electron renderer, such as cmd.exe, powershell.exe, bash, or node
Outbound network connections from FlyEnv to non-vendor destinations following markdown or chat rendering
Unexpected file reads or writes performed by the FlyEnv process under user profile paths

Detection Strategies

Inventory installed FlyEnv builds and flag any version below 4.18.0
Inspect markdown documents opened by FlyEnv for embedded HTML tags and JavaScript event handlers
Monitor process lineage for the FlyEnv Electron parent spawning shell or scripting interpreters

Monitoring Recommendations

Log FlyEnv process creation events and correlate child processes against a known-good baseline
Alert on filesystem access by FlyEnv outside its expected working directories
Capture network telemetry from FlyEnv and review unfamiliar domains contacted after opening untrusted content

How to Mitigate CVE-2026-69116

Immediate Actions Required

Upgrade FlyEnv to version 4.18.0 or later on all developer workstations
Avoid opening markdown files or importing AI chat transcripts from untrusted sources until patched
Restrict AI chat integrations in FlyEnv to trusted model endpoints only

Patch Information

The fix is delivered in FlyEnv 4.18.0 via pull request #810 and commit 68fd6d7. The patch introduces DOMPurify sanitization on all v-html sinks handling markdown and chat content. See the GitHub release v4.18.0, the issue report #809, and the VulnCheck XSS Advisory for full details.

Workarounds

Do not render markdown documents from untrusted contributors in vulnerable FlyEnv builds
Disable or avoid the AI chat feature until the upgrade is applied
Where feasible, run FlyEnv under a low-privilege OS account to limit filesystem exposure

bash

# Verify installed FlyEnv version and upgrade
flyenv --version
# Then download and install FlyEnv 4.18.0 or later from:
# https://github.com/xpf0000/FlyEnv/releases/tag/v4.18.0