cursor-ai

Pass

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: SAFENO_CODE
Full Analysis
  • Prompt Injection (SAFE): No instructions attempting to override agent behavior or bypass safety filters were detected. The role definition is strictly focused on AI editor expertise.
  • Data Exposure & Exfiltration (SAFE): No sensitive file paths or network exfiltration patterns were found. The skill does not perform any I/O operations.
  • Unverifiable Dependencies & Remote Code Execution (SAFE): The skill contains no code, package manifests, or remote script download patterns.
  • Privilege Escalation (SAFE): Mentions of 'sudo' and destructive commands (e.g., 'rm -rf') are provided solely as safety warnings for users to include in their own project rules.
  • Indirect Prompt Injection (LOW): The skill is designed to analyze and validate user-provided configuration files (.cursorrules).
  • Ingestion points: Processes .cursorrules and .mdc files.
  • Boundary markers: Absent in the skill instructions.
  • Capability inventory: No active scripts, shell access, or network capabilities exist in the skill.
  • Sanitization: Not applicable as the skill does not execute code or process data beyond natural language interaction.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 17, 2026, 06:13 PM