ask-matt

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

• Third-party content exposure detected (low risk: 0.10). The required runtime workflow described in SKILL.md/PHASE-BOUNDARIES.md centers on local repo issue/prototype management and “grilling” but does not specify any step that ingests outsider-authored free text from an external feed/queue or arbitrary web content without first selecting a specific item, so the runtime workflow as written does not expose the agent to attacker-posted poison via a broadly readable channel.