Cursor IDE Auto‑Executes Malicious Code from Poisoned Repositories
What Happened — Researchers disclosed a flaw in the Cursor AI‑assisted IDE that automatically runs code pulled from a Git repository. An attacker can poison a repo with malicious payloads, causing the IDE to execute the code on a developer’s machine without prompting. The vulnerability remains unpatched in the current release.
Why It Matters for Compliance & Audit Readiness
- The issue exemplifies a supply‑chain‑style control gap that SOC 2 continuous‑compliance programs are built to detect and evidence.
- Mapping this flaw to the CC6.1 – Change Management and CC7.1 – System Operations criteria demonstrates due‑diligence and provides audit‑ready proof that you monitor third‑party tool behavior.
- Verisq’s Control Mapping capability can automatically capture configuration drift and execution logs as continuous evidence for SOC 2 auditors.
Who Is Affected – SaaS development platforms, DevOps toolchains, and any organization that integrates AI‑assisted IDEs into its software‑development lifecycle (primarily TECH_SAAS).
Recommended Actions
- Inventory all AI‑assisted IDEs and assess their auto‑execution settings against SOC 2 CC6.1/CC7.1 controls.
- Enable explicit user confirmation for any code fetched from external repositories; log each execution event.
- Deploy continuous monitoring to capture IDE configuration changes and execution logs for audit evidence.
Technical Notes – The flaw stems from Cursor’s “auto‑run” feature that triggers on repository checkout, bypassing standard code‑review gates. No CVE has been assigned yet; the vulnerability is classified as a supply‑chain code‑execution risk. Source: Dark Reading