Loading ToolBench
Paid MCP accessibility evaluation server. AI agents pay per evaluation via x402 USDC micropayments. Evaluates design work against an expert-crafted accessibility rubric that goes beyond WCAG technical compliance.
Three tools with complete Zod schemas and detailed descriptions. Tool naming follows verb_noun pattern (list_capabilities, evaluate_typography, evaluate_accessibility). Descriptions are substantive (150-400 chars) and explain WHAT, WHEN, and pricing. Input schemas are comprehensive with proper type constraints, enums, and field descriptions. However, output schemas are not documented, LLMs cannot predict response structure. Error handling is absent; no guidance on payment failures, validation errors, or recovery paths. Tool annotations present (readOnlyHint, destructiveHint, idempotentHint) but incomplete, missing openWorldHint context for payment-gated tools.
Evaluates UI elements for accessibility issues beyond typography: touch target sizing, color-only information conveyance, focus indicator presence and contrast. Covers WCAG 1.4.1, 2.4.7, 2.4.11, 2.5.5, 2.5.8. COST: $0.01 USDC via x402 on Base-compatible EVM network per call.
Evaluates typography elements against a principled accessibility rubric. COST: $0.05 USDC via x402 on Base-compatible EVM network per call. Goes beyond what axe/Lighthouse/WAVE can check — evaluates design judgment, not just numeric compliance. Catches issues like contrast that passes WCAG 4.5:1 but fails visually due to thin font weight, body text that meets minimum size requirements but is still too small for comfortable reading, line heights that technically comply but impede readability for dyslexic users, extended all-caps or italic text that passes all AA criteria but impairs reading, text on gradient/image backgrounds where scanner sampling is unreliable, and heading sizes that are technically correct but visually indistinct from body.
Returns available evaluation tools, what they check, and their pricing. Call this first to understand what Axcess can evaluate and how much each evaluation costs. This tool is FREE. All evaluation tools require USDC payment on Base network. Returns: JSON with tool descriptions, pricing, and rubric categories.
Output schemas not documented. LLMs cannot predict response structure for evaluate_typography and evaluate_accessibility. Responses likely contain evaluation results, scores, and recommendations, but format is invisible to the agent.
No error handling guidance. Payment failures (insufficient USDC, network errors, x402 rejection), validation errors (invalid hex colors, out-of-range font sizes), and timeout scenarios lack recovery instructions. LLMs cannot self-correct or retry intelligently.
Payment mechanism not abstracted. Tool descriptions expose x402 protocol, USDC, Base network, and EVM details. Users/agents should not reason about blockchain infrastructure, wrap payment inside the tool and return simple success/failure.
Inferred effective spec: 2025-06-18+.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | C | 64 | 2025-06-18+ | v2 |
Tool annotations incomplete. evaluate_typography and evaluate_accessibility lack openWorldHint context, agents cannot determine if these tools can handle arbitrary design systems or only known patterns. readOnlyHint is correct but destructiveHint should clarify that payment is irreversible.
No pagination or result limits documented. evaluate_typography and evaluate_accessibility accept up to 50 elements per call. Response structure (single evaluation object? array of per-element results?) is undocumented, making it impossible to predict token cost or plan multi-batch workflows.