edge-candidate-agent

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

• Third-party content exposure detected (high risk: 0.90). The skill explicitly runs and ingests output from an external LLM CLI via --llm-ideas-cmd (see scripts/auto_detect_candidates.py and references/ideation_loop.md), and those returned hints are merged into the detection logic and scoring that determine which tickets are exported, so untrusted third-party output can materially change behavior.