payment-network
Pass
Audited by Gen Agent Trust Hub on Nov 16, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill fetches official tooling from established e-commerce and development providers.\n
- Downloads the Whop CLI installer from
whop.com.\n - Fetches the PowerShell installation script for the Whop CLI from the official
whopioGitHub repository.\n - References browser automation tools from the
vercel-labsGitHub organization.\n- [REMOTE_CODE_EXECUTION]: The skill provides commands to execute remote setup scripts for service integration.\n - Executes an installation shell script from
whop.comusingcurl | sh.\n - Executes a PowerShell script from the
whopioGitHub repository using theiexcommand.\n- [COMMAND_EXECUTION]: Leverages standard CLI tools and package managers to perform business operations.\n - Uses
npmandnpxto manage the@whop/clipackage.\n - Executes
whopcommands for authentication and account management tasks.\n- [INDIRECT_PROMPT_INJECTION]: The skill identifies a potential attack surface through the ingestion of external commerce data.\n - Ingestion points: Processes webhook payloads and commerce records as described in
references/commerce-record-conventions.mdandreferences/payment-state-and-safe-recovery.md.\n - Boundary markers: Explicitly instructs the agent to treat external page text and tool outputs as data rather than instructions in
references/browser-runtime-and-sessions.md.\n - Capability inventory: Involves the use of CLI tools, browser interaction, and account configuration writes.\n
- Sanitization: Implements verification of webhook signatures and immutable source record handling as defensive measures.
Audit Metadata