The Agent Skills Directory

[Prompt Injection] (SAFE): No malicious instructions intended to bypass safety filters or override agent constraints were detected. The content is educational in nature.
[Data Exposure & Exfiltration] (SAFE): No hardcoded credentials, sensitive file path references, or network exfiltration patterns were found.
[Obfuscation] (SAFE): The text is clear and readable with no signs of Base64 encoding, zero-width characters, or homoglyphs.
[Unverifiable Dependencies & Remote Code Execution] (SAFE): The skill does not include any package manager manifests or commands to download/execute remote scripts.
[No Code] (SAFE): This skill is categorized as safe because it consists entirely of documentation. While it includes code snippets as examples, these are illustrative and not part of an executable script environment.

prompt-engineering