spec-miner

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

• Third-party content exposure detected (high risk: 0.70). The skill’s required workflow uses tools like Glob/Grep/Read to ingest and quote “actual code evidence” from the target codebase (i.e., files the operating user did not author), which can include outsider-authored free text (source code comments/strings) that becomes LLM-readable context during analysis.