marketing-ops

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

• Third-party content exposure detected (high risk: 0.75). The skill’s runtime workflow can ingest outsider-authored free text from user-submitted Airtable Form inputs (e.g., “marketing request intake” descriptions/brief bodies) into LLM context for triage/AI drafting, and those form fields are authored by non-operating-user requesters.

MEDIUM W021: Hidden or invisible Unicode characters detected (potential obfuscation or prompt injection).

• Hidden Unicode characters detected (1 type(s) found)