meta-ads-api

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

• Third-party content exposure detected (high risk: 0.70). The skill explicitly calls the Meta Marketing API (e.g., Pre-flight checks: GET /me/permissions, GET /act_XXX, and checks like naming-collision and creative readiness that read campaign names/creatives from the account), so it ingests untrusted, user-generated third-party content from Meta which is then interpreted to decide whether to create/update/delete campaigns.

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

• Direct money access detected (high risk: 1.00). The skill explicitly integrates with the Meta Marketing API to create campaigns and adsets (POST /act_XXX/campaigns, POST /act_XXX/adsets) based on a campaign_spec that includes budget fields. It discusses account spend_caps, budget caps, and an activation step (either human or meta-ads-api activate <run_id>) that would enable live spending. This is a purpose-built tool for managing ad spend via a specific payment/spend API (Meta Ads), not a generic browser or HTTP tool — therefore it grants direct financial execution authority over advertising budgets.

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

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