find-twitter-influencers

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

• Third-party content exposure detected (medium risk: 0.30). The required workflow ingests outsider-authored free text from Exa’s contents.text by issuing an Exa search (POST https://exa.mpp.tempo.xyz/search) whose results are then parsed to extract Twitter handles, so a provider user can supply a query/similar URL that drives the agent to consume untrusted page text without first selecting a specific item.

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

• Direct money access detected (high risk: 1.00). The skill explicitly compiles steps into selat-pay calls and "pays the per-step price" (it invokes MPP/selat-pay and Circle Gateway settlement calls). The manifest and examples show POST/GET selat-pay commands that settle payments per step — i.e., the skill can initiate payment/settlement transactions. This is a direct financial execution capability.