battle

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

• Third-party content exposure detected (high risk: 0.90). The skill explicitly calls a /dogpile research skill (see the SKILL.md game loop "RESEARCH PHASE" and Task 14 / docs noting "Before each attack/defense, team can research strategies via dogpile" and config.DOGPILE_SKILL), which aggregates public third‑party providers (web, GitHub, YouTube, ArXiv, etc.), and those untrusted, user‑generated results are consumed and used to drive Red/Blue decisions (research → act → reflect), so they can materially influence tool use and next actions.