competitor-recon-weapon

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

• Third-party content exposure detected (high risk: 0.75). Yes: the runtime workflow described in this skill includes a public-artifact archaeology stage that probes and retrieves competitor public documents (e.g., /openapi.json, /.well-known/api-catalog, docs repos) which are outsider-authored free text/structured prose that can flow into the agent context for synthesis.