Creator Shield

Creator Shield is a continuous monitoring capability for platforms where creators receive payments. It's the "money in" half of risk management — every inbound payment to a creator_shield-enabled account triggers an evaluation against adverse media sources and Frame's proprietary risk indexes. If the creator surfaces as risky, the transaction is blocked, the account is restricted, and your platform is notified — before a chargeback hits, before a regulator asks questions.

Creator Shield is for the case where the creator is the source of risk to your platform, not the buyer. That's a distinction worth holding: standard fraud tooling protects against bad buyers; Creator Shield protects against bad sellers, where "bad" means engaging in activity that violates platform terms or creates legal or reputational exposure.

When to use it

Request creator_shield for accounts that fit this shape:

  • Creator platforms — adult content, performance, gaming, OnlyFans-style monetization. The risk is a creator engaging in prohibited activity off-platform that becomes your problem when funds flow.
  • Marketplaces with payout liability — gig platforms, freelance markets, anywhere a seller's behavior could trigger chargebacks, holds, or compliance scrutiny on your platform.
  • Any platform where ongoing seller monitoring matters more than one-time onboarding — KYC verifies identity at signup; Creator Shield watches what happens afterward.

If your platform's risk surface is primarily on the buyer side (people paying with stolen cards, refund abuse, etc.), that's Sonar territory, not Creator Shield. The two are complementary — Sonar is buy-side fraud; Creator Shield is sell-side risk.

How it works

Once creator_shield is active on an account, Frame runs evaluations continuously — there's no per-transaction call you make, no integration step beyond enabling the capability. The signal sources:

  • Adverse media — Frame queries news, legal records, regulatory filings, and public sources for coverage tied to illegal or prohibited activity attributable to the account holder. The check is identity-anchored (the KYC-verified name + DOB are what scope the search).
  • Frame's proprietary risk indexes — internal signals built from transaction patterns, platform history, and cross-merchant behavioral data on the Frame network. These catch patterns no single platform would see (e.g., the same creator's account behavior across multiple platforms).

When an inbound payment (card_receive or bank_account_receive) is initiated and a match surfaces:

  1. The transaction is blocked before it settles.
  2. The account is restricted — see the accounts concept for the restricted state model.
  3. Your platform is notified through Frame's standard event surface, with the account ID and a restriction reason.
  4. Frame's risk operations team reaches out directly to coordinate next steps.

The flow is invisible to creators during normal operation — there's no extra step for them, no consent screen beyond the standard KYC flow. Creators who never trip the signal never see it; flagged creators are blocked before they can withdraw.

KYC is the foundation

Creator Shield depends on a verified KYC record. Requesting creator_shield automatically includes kyc in the onboarding flow — you don't need to ask for both. Without verified identity, the adverse-media and risk-index checks have nothing to anchor against.

The pairing is implicit but worth understanding:

  • kyc verifies the creator is who they claim to be at the moment of signup.
  • creator_shield evaluates that verified identity continuously against external and Frame-internal risk signals.

KYC is a snapshot. Creator Shield is the ongoing watch.

Restriction handling

When Creator Shield flags an account:

  • The restricted state is set on the account; the restriction reason is visible in your Frame dashboard.
  • Inbound payments to the account stay blocked while the account is restricted.
  • Frame's risk operations team typically coordinates resolution directly with your platform — context-gathering, manual review, sometimes information from the creator themselves.

POST /v1/accounts/:id/unrestrict does exist on the API and can lift the restriction programmatically. In practice, you'll want to coordinate with Frame before lifting a Creator Shield restriction — the flag exists for a reason, and overriding it without context shifts the underlying risk back onto your platform. Treat the endpoint as a "last step after Frame ops sign-off," not a self-service auto-unblock.

What it doesn't do

Creator Shield is a risk-monitoring capability, not a one-stop content-moderation or platform-safety tool. It doesn't:

  • Replace your platform's own terms-of-service enforcement. Creator Shield catches a specific class of external risk; your platform still owns its content rules.
  • Provide real-time content review of what creators are posting. Frame doesn't see what creators publish — only what's surfaced through external news, legal, and risk-network sources.
  • Apply to buyer-side risk. Use Sonar for fraud signals on the money-in side.
  • Run on merchant-type or business-type accounts. It's an individual-account capability — the adverse-media check is identity-anchored on a person.

Gotchas

Symptom: creator_shield is stuck pending even after the creator completed onboarding. Why: KYC hasn't settled. Creator Shield can't evaluate until the upstream KYC capability is active. Fix: check kyc capability status first; if kyc is still pending (waiting on a document review) or disabled (verification failed), Creator Shield will follow suit.

Symptom: an account was restricted by Creator Shield and you don't see why in the dashboard. Why: some risk signals carry confidentiality constraints — Frame can surface the category of restriction reason but not the underlying source data. Fix: contact Frame's risk operations team; they'll walk through what they can share and the path to resolution.

Symptom: a creator who passed Creator Shield at onboarding gets flagged six months later. Why: that's the point — Creator Shield is continuous, not a one-shot. New adverse media, new patterns in the risk network, or new platform-level signals can flip an account from clear to flagged at any time. Fix: this is expected behavior; the alternative would be a one-time check that misses post-onboarding risk entirely.

Reference

For onboarding session creation, see POST/v1/onboarding_sessions.