automation-sandbox-post-copy-config-generate

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The prompt requires copying concrete SOP values into the emitted JSON and explicitly instructs embedding secrets found in URLs verbatim into the output, which forces the LLM to handle and output secret values directly.

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

• Third-party content exposure detected (medium risk: 0.30). SKILL.md step 1 locates and reads every supplied SOP source (including pasted excerpts, CSV/JSON/Markdown/plain text, and data-bearing PDFs/images) and then step 2–6 extracts actions/values from that outsider-authored content into the JSON output.