sfd-api

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

• Third-party content exposure detected (high risk: 0.90). The SKILL.md explicitly instructs the agent to fetch and ingest untrusted, user-generated content (e.g., POST /scrape for Instagram searches, POST /upload/url to pull arbitrary URLs, and POST /analyze with mediaUrl) in its required workflows, so third‑party content can be read and influence subsequent actions.

MEDIUM W009: Direct money access capability detected (payment gateways, crypto, banking).

• Direct money access detected (high risk: 1.00). The skill explicitly integrates a payment gateway and subscription/payments operations. POST /api_keys "Create an API key via Stripe checkout" (returns a Stripe checkout session URL) and GET /api_keys/status checks the Stripe session to retrieve the API key after payment. There is also POST /subscriptions to create a monthly paid subscription and several endpoints that deduct/refund credits (POST /accounts, POST /posts scheduling, POST /campaigns which "Credits deducted upfront", POST /campaigns/cancel refunds credits). These are specific payment-related functions (Stripe checkout and subscription/credit charging), not generic IO, so it grants direct financial execution capability.