The Agent Skills Directory

[SAFE]: The skill consists entirely of markdown documentation and code examples intended for educational use in prompt engineering.
[EXTERNAL_DOWNLOADS]: The skill references standard API libraries such as anthropic and google-genai. These are well-known, official packages from established AI providers used for demonstration purposes in the code snippets.
[DATA_EXFILTRATION]: No instances of sensitive data exposure or exfiltration were found. Code examples use safe placeholders like [paste text], [base64], and {id} for dynamic content.
[PROMPT_INJECTION]: While the skill discusses techniques like 'Chain-of-Thought' and 'ReAct', it does so in a descriptive and educational context. It contains no instructions directed at the agent to override its own safety protocols or system instructions.

prompt-engineering