The Agent Skills Directory

[PROMPT_INJECTION]: The skill includes documentation identifying common prompt injection patterns (such as instructions to ignore previous rules or reveal system prompts) specifically as examples of security risks for developers to mitigate.
[EXTERNAL_DOWNLOADS]: Technical documentation within the skill references common utility packages including zod, tiktoken, and openai as examples for implementing validation and token tracking in AI applications.
[SAFE]: The skill is primarily instructional and promotes robust security principles such as input sanitization, schema-based output validation, and the implementation of circuit breakers for API reliability.

ai-product