passthrough-api

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The skill's examples and required headers (x-pica-secret and x-pica-connection-key) show inline API/connection keys (e.g., curl -H "x-pica-secret: YOUR_API_KEY" and sample connection keys), meaning an agent using this skill would need to insert secret values verbatim into generated requests/commands, creating an exfiltration risk.

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

• Third-party content exposure detected (high risk: 0.90). This skill exposes the agent to untrusted, user-generated third-party content by executing arbitrary passthrough requests and returning underlying platform responses (see the Passthrough API endpoint /v1/passthrough/{path} and vault connections like /v1/vault/connections which can surface Gmail, Slack, HubSpot, Stripe, etc. data), which the agent is expected to read and interpret.

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

• Direct money access detected (high risk: 1.00). The passthrough API explicitly supports payment-platform connectors (example shown: Stripe) and provides a generic /v1/passthrough/{path} endpoint that can execute any action on connected platforms. That includes calling payment-gateway endpoints (e.g., Stripe) which can create customers, charges, refunds, etc. Because it exposes a direct integration to payment APIs (explicitly demonstrated), it grants direct financial execution capability.