The Agent Skills Directory

[SAFE]: The skill consists of educational documentation and templates for prompt engineering. All provided code snippets follow industry best practices for LLM interaction and do not contain malicious instructions.
[COMMAND_EXECUTION]: The file scripts/optimize-prompt.py is a local utility for A/B testing prompts. It uses standard Python libraries such as concurrent.futures and numpy to manage test runs and calculate metrics. It contains no suspicious subprocess calls or system-level modifications.
[EXTERNAL_DOWNLOADS]: The documentation references well-known and trusted AI libraries including langchain, anthropic, sentence-transformers, and scikit-learn. These references are standard for the described use case of prompt engineering and optimization.
[PROMPT_INJECTION]: While the skill demonstrates how to interpolate user data into prompts, it does not contain instructions that attempt to bypass safety filters or override agent behavior. Conversely, it includes patterns like 'Chain-of-Thought with Self-Verification' and 'Structured Outputs' which are designed to increase model reliability.

prompt-engineering-patterns