ma-playbook

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

• Third-party content exposure detected (high risk: 1.00). Flagged: the Multi-Agent Analysis (Step 2) Task prompts in SKILL.md explicitly instruct subagents to fetch and analyze public third-party sources (SimilarWeb, LinkedIn, G2/Capterra, Glassdoor, Blind, GitHub, job postings, public pricing pages), which are untrusted/user-generated and are used to drive due diligence, valuation, and follow-up actions.