The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes standard command-line tools such as grep, git, and npm audit to scan the local source directory. These commands are diagnostic in nature and limited to searching for security patterns (e.g., hardcoded secrets or outdated dependencies) within the user's project.
[DATA_EXFILTRATION]: While the skill searches for sensitive information like API keys and passwords, it does so within the context of a local audit. There are no instructions or network operations to transmit this discovered data to external servers.
[SAFE]: The functionality aligns perfectly with the skill's stated purpose of performing a security review. It follows best practices by encouraging the use of .env files and providing templates for vulnerability reporting without introducing any hidden or suspicious logic.

security-review