The Agent Skills Directory

[SAFE]: The skill implementation consists of shell scripts that perform read-only scans of project directories (src, app, pages, server, api, lib). It identifies strings related to personal data and telemetry libraries to alert the user of missing privacy documentation.
[SAFE]: Analysis of bin/check.sh confirms that it uses standard Unix utilities (grep, awk, jq) for pattern matching and JSON generation. It explicitly avoids reading sensitive credential files like .env or .env.production.
[SAFE]: No network operations, remote code downloads, or exfiltration patterns were detected in any of the provided files.
[SAFE]: The inclusion of bin/smoke.sh demonstrates good development practices, providing a suite of test cases to verify the scanner's logic against various project structures.
[SAFE]: The skill follows the principle of least privilege, requiring only read access to project source code and utilizing a separate mechanism for saving build artifacts.

privacy-check