Ovra Payments

Use this skill when the agent needs to buy something online, pay for a service, complete a checkout, manage spending budgets, or track purchases. Enables sec...

This skill appears coherent for agent-driven payments, but before installing: 1) Verify the Ovra service and endpoint (https://api.getovra.com) and confirm you trust the provider; 2) Use a scoped or sandbox API key for initial testing (and confirm what 'no spending limits' for sandbox keys actually means); 3) Limit the OVRA_API_KEY's permissions if possible and rotate/revoke keys regularly; 4) Require agent approval workflows or human-in-the-loop for real-money transactions rather than allowing fully autonomous charges; 5) Be cautious with the receipt upload feature — ensure the agent is only allowed to read and upload transaction receipts and not arbitrary local files; 6) Monitor billing/transactions and enable alerts; and 7) If you need clarification (GDPR/data retention, what data Ovra stores, audit logs), ask the vendor before granting production credentials.

Type: OpenClaw Skill Name: ovrapay Version: 1.0.0 The skill provides a legitimate payment integration for AI agents via the Ovra platform. It uses the Model Context Protocol (MCP) to communicate with a centralized API (api.getovra.com) and implements security measures such as tokenized credentials to prevent the agent from accessing raw card data. The instructions in SKILL.md clearly define operational boundaries, including requirements for human approval for high-risk actions like policy changes or large transactions, and no evidence of malicious intent, data exfiltration, or unauthorized execution was found.