The Agent Skills Directory

[SAFE]: No security issues were detected during the analysis of this skill.
[PROMPT_INJECTION]: The instructions focus entirely on IDE configuration and shortcut management. There are no attempts to bypass safety filters or override agent behavior.
[DATA_EXFILTRATION]: The skill does not access sensitive files, environment variables, or hardcoded credentials. No network exfiltration patterns were observed.
[REMOTE_CODE_EXECUTION]: The skill does not download or execute remote scripts. External links point to official documentation for Cursor, VS Code, and the Vim extension, which are well-known and trusted services.
[OBFUSCATION]: All content is in plain text. No Base64, zero-width characters, or homoglyph attacks were identified.
[COMMAND_EXECUTION]: While the skill's manifest allows the Bash tool, the provided instructions and reference files do not contain any dangerous shell commands or automated execution patterns.
[INDIRECT_PROMPT_INJECTION]: The skill documents the processing of keybindings.json files, which is a standard IDE configuration task and does not introduce a significant injection surface for malicious instructions.

cursor-keybindings