HighVulnerability

GHSA-299v-8pq9-5gjq: New API has Potential XSS in its MarkdownRenderer component

Source: GitHub Advisory(opens in a new tab)
Identifiers: CVE-2026-25802GHSA-299v-8pq9-5gjq
Published: February 23, 2026
Record updated: October 9, 2026

Affected: github.com/QuantumNous/new-api < 0.10.8-alpha.9
Fixed in: 0.10.8-alpha.9
released February 6, 2026, 17 days before disclosure
Known exploitation: Not listed in the CISA Known Exploited Vulnerabilities catalog at the last check.
EPSS: 0.3%

Summary

The New API project has a cross-site scripting flaw in its MarkdownRenderer.jsx component, where lines 212-231 use dangerouslySetInnerHTML to preview HTML that the model generates. A model response containing a script tag can run in the playground, and because chat records are stored, the script re-triggers whenever the page is reopened. The advisory notes the impact applies if the model API is altered to output unintended code.

Mitigation

The preview may be placed in an iframe sandbox, and dangerous HTML strings should be purified before rendering.

Details

Scores and exploitation

EPSS: 0.3%Estimated probability that this CVE is exploited in the next 30 days, from the FIRST Exploit Prediction Scoring System. Read on November 4, 2026.

Weaknesses and techniques

AI threat technique (MITRE ATLAS): AML.T0054

Affected

Packages: github.com/QuantumNous/new-api@< 0.10.8-alpha.9 (fixed: 0.10.8-alpha.9)

Classification

Attack type: Jailbreak,
Other
Attack sophistication: Trivial
AI component targeted: API
Impact: Integrity,
Confidentiality
Labels: security

How this record was produced

Summaries, mitigations and labels on this page are generated automatically from the source. The linked source is authoritative.

Original source: https://github.com/advisories/GHSA-299v-8pq9-5gjq

First tracked: February 23, 2026 at 07:00 PM ET, from GitHub Advisory Database

Classifier: claude-haiku-5-5, prompt v4

Classifier confidence: 80% (self-assessed)

Summary

Mitigation

Related items