Loading ToolBench
MCP server that enables AI agents to create Stripe products, manage checkout sessions, and process payments with Visa TAP cryptographic signatures
Four tools with basic descriptions and partial schemas. All tools have descriptions (10-200 chars range), but parameter documentation is inconsistent. Input schemas present but lack formal type constraints (no enums, ranges, or validation hints). Error handling returns plain strings without recovery guidance. No tool annotations (readOnlyHint, destructiveHint, idempotentHint). Naming is verb-forward but descriptions lack WHEN/WHY context. Output schemas undocumented, LLMs cannot predict response structure for chaining.
Expires an open Stripe Checkout Session immediately. Use this to cancel a pending purchase or deactivate a checkout URL.
Simulates an AI agent completing a transaction via the Merchant Storefront Gateway. Automatically generates real-time RFC 9421 cryptographic signatures for Visa TAP.
Creates a Stripe Checkout Session for a specific price ID. Returns the URL where the user or agent can complete the payment.
Creates a new product and price inside the Stripe sandbox. Takes price in cents (e.g., 2000 for $20.00). Returns the Price ID.
No output schemas documented. LLMs cannot predict response structure (fields, types, nested objects) for downstream tool chaining or data extraction.
Parameter descriptions lack format/constraint hints. 'price_id' and 'session_id' have no format guidance (length, pattern, prefix). 'quantity' has no min/max bounds. LLMs may pass invalid values.
Error responses are plain strings without recovery guidance. 'Failed to create product: {error}' tells LLM nothing about retry eligibility or next steps. No error classification (retryable vs fatal).
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | C | 63 | 2026-07-28+ | v2 |
No tool annotations (destructiveHint, idempotentHint). create_merchant_checkout and complete_sandbox_payment are WRITE operations but lack destructiveHint. LLMs cannot distinguish safe from risky tools.
Descriptions lack WHEN/WHY context. 'Creates a Stripe Checkout Session' does not explain when to call this vs complete_sandbox_payment, or what the user sees next. LLMs may conflate similar tools.