The Agent Skills Directory

[COMMAND_EXECUTION]: The file references/recovery-strategies.md includes shell commands for resetting the Cursor IDE state, such as rm -rf ~/Library/Caches/Cursor/ and rm -rf ~/.cursor/extensions/. These are documented as manual 'Reset Options' for recovery and are not configured for autonomous execution by the agent.
[DATA_EXFILTRATION]: No exfiltration patterns or unauthorized network operations were detected. The skill explicitly promotes security best practices in references/security-pitfalls.md, advising the use of .cursorignore to protect sensitive files like .env and recommending against hardcoding credentials.
[PROMPT_INJECTION]: The skill facilitates responses to user queries regarding 'cursor pitfalls,' which creates an indirect prompt injection surface.
Ingestion points: User queries triggering the skill's keyword-based patterns.
Boundary markers: Not explicitly defined in the provided instructions.
Capability inventory: The skill has access to broad Bash execution, File Read, Write, and Edit tools.
Sanitization: No specific input sanitization or validation logic is implemented for user-supplied data before it is processed by the agent in the context of these capabilities.
[EXTERNAL_DOWNLOADS]: The skill references official resources for the Cursor IDE, including documentation, community forums, and Discord. No unverified third-party downloads or remote code execution patterns were identified.

cursor-known-pitfalls