---
title: "CVE-2025-54135: Cursor Agent is vulnerable to prompt injection via MCP Special Files"
url: https://erroneo.us/security/cve/CVE-2025-54135/
description: "CVE-2025-54135 (cursor cursor): Cursor is a code editor built for programming with AI. Cursor allows writing in-workspace files with no user approval in…"
type: vulnerability-record
section: "Security Digest"
date_published: 2025-08-05
date_modified: 2026-10-21
publisher: erroneo.us
author: erroneo.us Security Desk
cve: "CVE-2025-54135"
kev: false
cvss: 8.6
epss: 0.0181
license: All rights reserved. Quote with attribution and a link.
---

# CVE-2025-54135: Cursor Agent is vulnerable to prompt injection via MCP Special Files

**Signal:** Normal patch cycle. No exploitation signal; fix it in your next routine update if you are affected. (Computed from CISA KEV, EPSS and CVSS; not an editorial verdict.)

## Should you act?

- No confirmed exploitation. It is not in CISA’s KEV catalog and CISA’s SSVC assessment records no known exploitation. That can change quickly after a public write-up; re-check if you are affected.
- EPSS 1.8% , percentile 78: FIRST’s model estimates a 1.8% chance of exploitation activity in the next 30 days, higher than 78% of all scored CVEs (moderate). EPSS is a probability, not a verdict.
- Needs an account (low (any account)). Exposure depends on who can log in.
- Editorial verdict: Most can relax. Cursor CurXecute: Prompt Injection to RCE via mcp.json — Cursor auto-updates and both bugs were fixed by mid-2025; the lasting takeaway is to treat any agent that can write its own tool config as a code-execution path.

## What it is

Cursor is a code editor built for programming with AI. Cursor allows writing in-workspace files with no user approval in versions below 1.3.9, If the file is a dotfile, editing it requires approval but creating a new one doesn't. Hence, if sensitive MCP files, such as the .cursor/mcp.json file don't already exist in the workspace, an attacker can chain a indirect prompt injection vulnerability to hijack the context to write to the settings file and trigger RCE on the victim without user approval. This is fixed in version 1.3.9.

## Affected products (CVE record)

- cursor cursor: < 1.3.9

## Severity

- GitHub_M CVSS 3.1: 8.6 `CVSS:3.1/AV:N/AC:H/PR:L/UI:N/S:C/C:H/I:H/A:H`

## Exploitation

- CISA KEV: not listed (no confirmed in-the-wild exploitation known to CISA).
- CISA SSVC assessment: exploitation none , automatable no , technical impact total .
- FIRST EPSS (model of 2026-10-21): 1.81% probability of exploitation activity in the next 30 days; percentile 78.

## Sources

- [CVE.org record CVE-2025-54135](https://www.cve.org/CVERecord?id=CVE-2025-54135)
- [NVD entry CVE-2025-54135](https://nvd.nist.gov/vuln/detail/CVE-2025-54135)
- [github.com](https://github.com/cursor/cursor/security/advisories/GHSA-4cxx-hrm3-49rm)

---
Source page: https://erroneo.us/security/cve/CVE-2025-54135/ · Security Digest: https://erroneo.us/security/ · More: https://erroneo.us/llms.txt