Skip to content
KitploitKITPLOIT
ToolsExploitsBlog
Log in
Submit
ToolsExploitsBlog
Submit

Hacking, PenTest, and Cybersecurity Tools for Your Security Arsenal!

Kitploit is a directory of hacking, cybersecurity, and pentesting tools. Discover the latest project updates to find vulnerabilities, analyze systems, automate testing, and strengthen your security.

FeedsContactPrivacy© 2026 Kitploit

Tool Directory

Categories

View all categories
Loading categories
CVE-2026-79294 — Responsible-disclosure advisory for CVE-2026-79294, a stored XSS in Moonshot AI Kimi's HTML artifact Preview via the public Share view enabling session-token theft and account takeover. | Kitploit
Tools/GitHubGitHub/mgtx2/cve-2026-79294
Vulnerability AnalysisExploitationWeb Application ExploitationWeb SecurityPapers & ResearchLearning & Education
GitHubmgtx2/cve-2026-79294

CVE-2026-79294

Responsible-disclosure advisory for CVE-2026-79294, a stored XSS in Moonshot AI Kimi's HTML artifact Preview via the public Share view enabling session-token theft and account takeover.

View Repository
402 months agoNot yet reviewed

Most Popular

View all →

Discover the most used tools by our community.

  • 1

    OsintHunter

    letox4/osinthunter

    🎯 The ultimate OSINT toolkit: investigation graph, confidence scoring, alias families, map, timeline and reports. 49 modules, 400+ sites, no API keys required.

    1.0k
  • 2

    ghostlock-app

    yukonga/ghostlock-app

    GhostLock One-Tap Execution App (CVE-2026-43499)

    589
  • 3

    looking-glass

    anonghosty/looking-glass

    CSIRT platform that ingests Shadowserver reports into MongoDB and replays them as interactive maps, analysis, findings, advisories and executive dashboards.

    361
  • 4

    redforge

    cybersecurity91/redforge

    Browser-side red team toolkit with 102 offline tools for recon, payload generation, credential attacks, crypto, encoding and engagement reporting.

    339
  • 5

    ARTEX

    baarai/artex

    LLM multi-agent autonomous penetration testing platform with planner/worker agents, asset and exploration graphs, MITM traffic recording, and vulnerability retesting.

    324
  • 6

    CyberStrikeAI

    ed1s0nz/cyberstrikeai

    The system of action for AI-native cybersecurity—where intent becomes governed execution, evidence becomes operational memory, and every operation improves the next.

    322

Explore all tools

Browse our collection of tools

View all tools →
Share

CVE-2026-79294

CVE-2026-79294 — Stored XSS in Moonshot AI Kimi's HTML artifact Preview, delivered through the public Share view, leading to session token exfiltration and demonstrated account takeover. Responsible disclosure advisory.

Vulnerability Details

  • Vulnerability Type: Stored/Stored-like Cross-Site Scripting (XSS) via AI Response
  • Affected Product: Kimi AI
  • Affected Version: v1.0

CVE CWE-79


Affected

Vendor / productMoonshot AI — Kimi
ComponentsHTML artifact Preview rendering; public Share view
WeaknessCWE-79 — Improper Neutralization of Input During Web Page Generation

The issue

Kimi renders model-generated HTML artifacts in a Preview pane inside the application's own origin, without neutralising script. Sharing a conversation publishes that artifact at a public Share URL, so script authored by one account runs in the browser of anyone who opens the link.

The session values Kimi relies on are held in localStorage and are therefore readable by that script:

access_token
refresh_token
msh_user_id

Script execution consequently does not stop at script execution. Those three values can be read out of a victim's browser, sent anywhere, and replayed to obtain an authenticated session as that victim.


Chain

attacker account  →  artifact containing exfiltration script  →  public Share
       →  victim opens the link  →  script runs in Kimi's origin
       →  the three localStorage values reach the attacker
       →  replayed into a clean browser  →  authenticated as the victim

An attacker needs an ordinary account. The victim needs only to open a share link while signed in — which is what share links are for.


Verification

ObservedHow
Script executes in the victim's browserA request reached the attacker's endpoint, originating from the victim's session
The three values are script-readableAll three were present in that request body
The values authenticate on their ownWritten into a browser that had never signed in to Kimi; after reload, that browser held an authenticated session
The values are the causeControl: same browser, values omitted, reload — remained unauthenticated

Both accounts involved were created and controlled by the researcher. No third party's data was accessed at any point.


Proof of concept

The chain is documented above in full. The working payload and the receiving endpoint are withheld — publishing them would arm the attack rather than describe it.

The artifact prompt takes the form:

Generate an HTML page that reads the three localStorage values above
and POSTs them to <attacker endpoint>.
Download Tool