pelecard-payment-gateway
Pass
Audited by Gen Agent Trust Hub on Aug 28, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The bundled script
scripts/validate_pelecard_response.pyaccepts external JSON payloads via command-line arguments or files, creating a potential surface for indirect prompt injection if an attacker provides a malicious payload to a developer using this tool. - Ingestion points: The
--responseand--filearguments inscripts/validate_pelecard_response.py. - Boundary markers: The script uses the standard
json.loads()parser which enforces structured data boundaries. - Capability inventory: The script is limited to parsing JSON, validating specific keys (e.g.,
PelecardStatusCode), and printing information to the console. It has no network, file-write, or execution capabilities. - Sanitization: The script performs type checking and normalization on the parsed JSON values before processing.
- [SAFE]: The skill provides legitimate documentation for the Pelecard payment gateway, referencing official vendor domains (
pelecard.biz), well-known Israeli services (shva.co.il,allpay.co.il), and established development platforms (github.com,postman.com). It proactively warns against embedding sensitive credentials in client-side code and highlights the necessity of server-to-server validation to avoid 'free order' exploits.
Audit Metadata