The Agent Skills Directory

[SAFE]: The skill serves as a documentation and template resource for prompt engineering. It includes examples for few-shot learning, chain-of-thought, and structured output using standard Python libraries.
[DATA_EXPOSURE]: No hardcoded credentials, API keys, or sensitive file paths were found. The code snippets use generic placeholders for integration.
[EXTERNAL_DOWNLOADS]: The skill references standard, well-known libraries such as pydantic and openai. No suspicious external URLs or remote script downloads are present.
[PROMPT_INJECTION]: While the skill contains instructions on how to structure prompts, it does not contain patterns intended to bypass AI safety guidelines or override system instructions.
[INDIRECT_PROMPT_INJECTION]: The skill identifies potential attack surfaces by providing templates that ingest untrusted data (e.g., {code}, {error_log}). However, it also demonstrates best practices such as using XML tags for boundary isolation and provides checklists that warn against prompt injection vulnerabilities.

prompt-engineering