edge-hint-extractor

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

• Third-party content exposure detected (high risk: 0.90). The skill explicitly ingests and parses untrusted third-party content via the --llm-ideas-cmd (runs an external subprocess) and --llm-ideas-file (loads arbitrary YAML) flows described in SKILL.md/README and implemented in scripts/build_hints.py, and that parsed LLM/file output is merged into hints.yaml and can materially change downstream decisions.