oneshot-prompt-generator

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 0.90). The prompt explicitly tells the model to "preserve exact copy and literals" and to "embed the material needed for a self-contained result" when access won't carry over, which can force the LLM to include secret values verbatim in its output.

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (medium risk: 0.30). SKILL.md directs the agent to inspect the user-supplied resources and treat all extracted text (including any embedded instructions) as untrusted evidence, meaning outsider-authored free text can enter via the supplied “documents/media/code/page” evidence and be ingested to produce the final prompt.