The Agent Skills Directory

[SAFE] (SAFE): The skill is purely educational and provides standard patterns for authentication implementation. No malicious code, obfuscation, or data exfiltration attempts were found.
[CREDENTIALS_UNSAFE] (SAFE): Example files like examples/auth-flows.json use standard placeholders such as 'YOUR_CLIENT_ID' and 'YOUR_SECRET', along with dummy JWT tokens for demonstration purposes only.
[COMMAND_EXECUTION] (SAFE): The included scripts (scripts/jwt-validator.py, scripts/session-checker.sh) are low-risk local utilities for decoding tokens and checking cookie flags. They do not involve remote code execution or dangerous subprocess calls.
[EXTERNAL_DOWNLOADS] (SAFE): No external dependencies or remote script downloads are performed. All scripts and references are contained locally within the skill.

authentication-patterns