Obtain Link purchase approval and complete checkout with server-side credential entry in the existing browser.
80%
Does it follow best practices?
Run evals on this skill
Adds up to 20 points to the overall score
Medium
Suggest reviewing before use
Fix and improve this skill with Tessl
tessl review fix ./apps/templates/bell-stack/agent/extensions/link/skills/create-payment-credential/SKILL.mdThe skill is specifically designed for direct financial operations, giving the agent the ability to move money or execute financial transactions — such as payment processing, cryptocurrency operations, banking integrations, or market order execution.
The skill operates real purchase checkout (direct financial operation: charging/purchasing via Link checkout) outside of operator-enabled sandbox mode, where "Responses marked test:true" are the exception rather than the default, so a live, real-money purchase route is in scope. The checkout tool enforces independent control: approval is created and retrieved via separate Link approval endpoints, and the checkout tool itself "rechecks approval and the merchant/amount/currency" and "retrieves credentials privately" before submission (service/execution-layer enforced, not merely agent-followed instructions), with submission blocked on denial/cancellation/expiry and mismatched merchant/amount/currency. This constitutes qualifying independent authorization for the direct operation, placing it at Medium rather than High.
The skill exposes the agent to untrusted, user-generated content from public third-party sources, creating a risk of indirect prompt injection. This includes browsing arbitrary URLs, reading social media posts or forum comments, and analyzing content from unknown websites.
The skill workflow involves browser checkout on merchant pages, which can expose the agent to untrusted merchant content and payment flows.
If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.