integrate-payment-gateway
Pass
Audited by Gen Agent Trust Hub on Oct 31, 2026
Risk Level: SAFEREMOTE_CODE_EXECUTIONCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [REMOTE_CODE_EXECUTION]: The skill instructs the agent to use 'npx mppx@latest validate' for validating Machine Payments Protocol (MPP) integrations. This command downloads and executes an external package from the NPM registry at runtime.
- [COMMAND_EXECUTION]: The documentation includes validation gates that require running local scripts such as 'python3 core/scripts/validate-all.py' and 'validate-rules.py'.
- [INDIRECT_PROMPT_INJECTION]: The skill represents an indirect injection surface as it is designed to ingest and process untrusted data from external payment webhooks and API responses. * Ingestion points: The skill uses the 'fetch' tool and implements webhook handlers for various payment providers. * Boundary markers: There are no explicit instructions to use delimiters or ignore instructions within the processed payment payloads. * Capability inventory: The skill utilizes 'write_file', 'create_file', 'run_tests', 'run_build', and 'run_linter', which could be targeted if an attacker successfully injects instructions into a webhook payload. * Sanitization: The skill follows best practices by recommending HMAC signature verification and strict schema validation (ASI05), which mitigates but does not fully eliminate the indirect injection risk.
Audit Metadata