account-scoring

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

• Third-party content exposure detected (medium risk: 0.30). The workflow plays/score-accounts.ts feeds an outsider-supplied input.name and input.domain into the accountScorer agent prompt (prompt: \Score the account ${input.name} (${input.domain}).``) and then the agent’s system prompt/LLM can ingest that free text at runtime before scoring and writing the rationale to the CRM.