AI Science Tracker

Taskflow Agent finds MIME-type spoofing XSS in Docmost

Docmost · AI involvement attributed to GitHub Security Lab Taskflow Agent

Finding and attribution

Docmost trusts the uploaded MIME type when serving files inline while deciding download behavior from the filename extension; a file named as an image but typed as HTML can execute script in the site origin.

Product: Docmost. AI system: GitHub Security Lab Taskflow Agent. Team / organization: GitHub Security Lab.

CVE identifiers: CVE-2026-33193.

AI and human contribution (source account): GHSL explicitly attributes discovery to Taskflow Agent and manual verification by Peter Stöckli and Man Yue Mo; report date 2026-02-17, fix date 2026-02-20.

Human involvement is only described where the source account establishes it; unknown involvement does not mean none. Vendor confirmation of a flaw does not confirm every claim about AI contribution.

Discovery context: Public advisory documents the vulnerability and a remediation; it does not claim exploitation in the wild.

Remediation status: fix documented. GHSL says the fix was made on 2026-02-20 and included in the next release; the advisory page does not state its version number.

A source-documented fix does not establish installed patches or current safety. Source checking does not independently reproduce the finding or establish observed exploitation.

Disclosed 24 Apr 2026. AI attribution dated 24 Apr 2026. Sources checked 14 Nov 2026; these are separate dates, not a live installation check.

Recorded CVSS: Medium · highest recorded score 4.6. Documented technical effects: code execution.

Severity, consequences and exploitation evidence

Case grouping uses the highest recorded CVSS score, with its version-specific severity band. It is descriptive, not a combined or editorial risk index. All recorded assessments remain below; scores from different versions and sources may disagree. CVSS specifications define version-specific scoring and severity bands.

CVE-2026-33193: code execution.
Prerequisites / affected scope: A user who can upload MIME-spoofed content can cause stored script to execute in browsers of users viewing it; record describes potential compromise of users/data..
Exploitation evidence: The cited source does not establish exploitation in the wild.. Impact source · checked 16 Nov 2026.

Documented fix: GHSL says the fix was made on 2026-02-20 and included in the next release; the advisory page does not state its version number.

Original evidence

Download this source record (JSON)

Open this case in the interactive tracker · Browse all static security cases