payram-openclaw-integration
Fail
Audited by Gen Agent Trust Hub on Oct 14, 2026
Risk Level: HIGHEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: Fetches a setup script from the vendor's official domain at
payram.com. - [REMOTE_CODE_EXECUTION]: Provides instructions to execute the vendor's setup script using bash process substitution:
bash <(curl -fsSL https://payram.com/setup_payram_agents.sh). - [INDIRECT_PROMPT_INJECTION]: The skill is designed to process external data from payment webhooks and messaging platforms, creating a potential surface for indirect prompt injection.
- Ingestion points: Webhook payloads from payment events, customer information in payment creation requests, and inbound messages from chat platforms like WhatsApp, Telegram, and Discord.
- Boundary markers: The instructions do not specify the use of clear delimiters or guardrails when interpolating untrusted data into agent prompts.
- Capability inventory: The skill provides tools for generating code snippets, project scaffolding, and managing local workers (e.g.,
restart_payram_worker). - Sanitization: The skill documentation recommends verifying webhook authenticity using HMAC-SHA256 signatures, though it does not explicitly detail input sanitization for user-provided strings.
Recommendations
- HIGH: Downloads and executes remote code from: https://payram.com/setup_payram_agents.sh - DO NOT USE without thorough review
Audit Metadata