The Agent Skills Directory

[Prompt Injection] (SAFE): No instructions attempting to override agent behavior or bypass safety guidelines were found. The troubleshooting section mentions safety triggers as a conceptual topic.
[Data Exposure & Exfiltration] (SAFE): No sensitive file paths, hardcoded credentials, or unauthorized network operations were detected. The OpenAI client initialization uses standard placeholders/default environment variable logic.
[Indirect Prompt Injection] (LOW): The skill provides templates that interpolate user-controlled data (e.g., {user_review}, {document}) directly into prompts. This pattern is a known vector for indirect injection in LLM applications; however, in this context, it is presented solely as an educational example for prompt design.
[Obfuscation] (SAFE): No encoded strings, homoglyphs, or hidden Unicode characters were identified.
[Unverifiable Dependencies & Remote Code Execution] (LOW): The code snippets reference the openai library, which is a standard industry package. No untrusted remote scripts or piped bash executions are present.

prompt-engineering