The Agent Skills Directory

[PROMPT_INJECTION]: Indirect prompt injection surface detected. The skill retrieves data from the Stripe API (such as customer names, product descriptions, and metadata) which can be controlled by external parties and displays it to the agent without sanitization.
Ingestion points: scripts/customers.rb, scripts/customer.rb, scripts/products.rb, scripts/invoices.rb and other listing/detail scripts.
Boundary markers: Absent. The skill does not use delimiters or instructions to the agent to disregard instructions within the data.
Capability inventory: The skill possesses the ability to read and write files and perform network operations via the included Ruby scripts.
Sanitization: Absent. The scripts print raw API response data directly to the output.
[DATA_EXFILTRATION]: The skill manages sensitive credentials.
Evidence: The skill stores the Stripe Secret Key in ~/.config/stripe/token and sets file permissions to 0600 in scripts/save_token.rb.
Evidence: scripts/save_token.rb accepts the API key as a command-line argument, which may be visible in system process logs (e.g., via ps).

stripe