copy-json

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). This skill extracts arbitrary JSON string fields into a temporary markdown file and sends that file to the copywriter skill (and later shows before/after text), so any API keys/passwords or secrets present in those fields would be passed through and could be output verbatim by the LLM, creating an exfiltration risk.

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

• Third-party content exposure detected (high risk: 0.75). Outsider-authored free text can enter the LLM context via the runtime-provided JSON file at FILE_PATH (Step 1 reads and extracts string values, writes them into a temp markdown, then Step 2 invokes the copywriter skill with that temp markdown content, which the LLM reads).