HighVulnerabilityLLM-specific
CVE-2025-64107: Cursor editor command execution via backslash path manipulation on Windows
- Source
- NVD(opens in a new tab)
- Identifier
- CVE-2025-64107
- Published
- Record updated
- Known exploitation
- Not listed in the CISA Known Exploited Vulnerabilities catalog at the last check.
- EPSS
- 0.4%
Summary
CVE-2025-64107 affects Cursor, a code editor built for programming with AI, in versions 1.7.52 and below. Cursor checks path manipulation that uses forward slashes, such as ./.cursor/./././././mcp.json, and requires human approval before completing the operation. The same manipulation with backslashes was not detected, so an attacker who already had prompt injection or another level of control could overwrite sensitive editor files without approval on Windows machines, which may lead to RCE.
Mitigation
Fixed in version 2.0.
Related items
- InfoTop agent skill security resources — November 2026Similar attack · Adversa AI Blog
- InfoHow to Turn AI Red Team Findings Into Runtime ControlsSimilar attack · Straiker Blog
- MediumDocker Sandboxes vulnerability lets AI agents access sensitive files outside their sandbox (CVE-2026-103916) | Endor LabsSimilar attack · Endor Labs Blog
- MediumFree the Model, Sell the Gateway: Prompt Adware hidden in Jailbreak BundlesSimilar attack · Check Point Research
- InfoRight Tool, Wrong Parameters: Defending LLM Agents Against Parameter PoisoningSimilar attack · Arxiv (cs.CR + cs.CL + cs.LG)