skills/stripe/link-cli/link-balance/Gen Agent Trust Hub

link-balance

Pass

Audited by Gen Agent Trust Hub on Nov 17, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [Automated Payment Management]: The skill facilitates purchases without manual user approval using the Link balance feature. It incorporates robust safety checks, instructing the agent to verify user-defined limits and account balances before each transaction to prevent unintended spending.
  • [Credential Safety]: The instructions provide guidance on preventing the exposure of sensitive credentials, specifically advising the agent not to include secret keys in generated setup URLs and to treat auth tokens as sensitive information.
  • [Indirect Prompt Injection Surface]: As the agent is expected to read merchant page content to facilitate purchases, the skill explicitly addresses the risk of indirect injection. It warns the agent to base purchase decisions solely on the user's expressed goals and instructions, ignoring any suggestions or prompts found within untrusted external sources like merchant ads or page content.
  • [Vendor Resource Usage]: The skill utilizes the @stripe/link-cli package and directs users to link.com. These are legitimate resources provided by the vendor for the intended functionality of the skill.
Audit Metadata
Risk Level
SAFE
Analyzed
Nov 17, 2026, 09:56 PM