Loading ToolBench
Trading-card intelligence MCP server. Identify cards, look up recent comparable sales, summarize market value, chart price history, and compare seller economics across eBay, Pulltrader, and other marketplaces.
Two well-defined tools with comprehensive schemas, clear descriptions, and proper parameter documentation. Both tools follow verb_noun naming (compare_*, calculate_*) and include detailed descriptions (200+ chars) explaining WHAT, WHEN, and constraints. Schemas are complete with type definitions, enums, and ranges. However, output schemas are not documented in the source, and error handling guidance is minimal. No tool annotations (readOnlyHint/destructiveHint) despite both being read-only operations.
Given a target take-home amount, compute the per-item sale price a trading-card seller must list at to net that amount on ONE selling method, after fees — eBay (estimated), a Pulltrader selling method (marketplace, Fulfilled by Pulltrader, branded storefront, in-person POS), or an estimated competitor marketplace (TCGplayer, Mana Pool, Misprint, Fanatics Collect, Goldin). If acquisition_cost is supplied, the target is treated as net profit (payout minus what you paid); otherwise it is the take-home payout. Use this when a seller asks 'what do I need to list this at to walk away with $X', 'to net/profit $X after fees', or 'to break even'. Calculations are deterministic and use dated fee schedules. Do NOT use this to look up a card's market value or recent sales (this tool does not price cards), and do NOT use it for non-trading-card categories. Present eBay and competitor figures as estimates, never as guaranteed proceeds.
Compare estimated fees and the net amount a trading-card seller keeps when selling the SAME card across eBay (estimated), Pulltrader selling methods (marketplace, Fulfilled by Pulltrader, branded storefront, and in-person POS), and other marketplaces (TCGplayer, Mana Pool, Misprint, Fanatics Collect, Goldin — estimated fixed-price/Buy Now seller fees). Use this when a seller asks what they would keep/net/take-home on a sale, how fees compare between platforms, or which method leaves them with more money. Calculations are deterministic and use dated fee schedules. Competitor marketplaces are off by default; include them via the `methods` field. Only fixed-price seller fees are modeled — auction formats (hammer price, buyer's premium, negotiated consignment) are not. Do NOT use this to look up a card's market value or recent sales (this tool does not price cards), and do NOT use it for non-trading-card categories. Present competitor and eBay figures as estimates, never as guaranteed proceeds, and never claim one platform is universally cheapest.
Output schemas not documented. Tool descriptions explain inputs but do not specify what fields/structure the response contains. LLMs cannot plan downstream operations without knowing response shape.
Tool annotations missing. Both tools are read-only (no side effects), but lack readOnlyHint annotation. This prevents clients from optimizing caching, retry logic, or safety checks.
Error recovery guidance absent. Tool descriptions warn against misuse (e.g., 'Do NOT use this to look up a card's market value') but do not guide LLMs on what to do if they make that mistake or if the tool fails.
Inferred effective spec: 2025-06-18+.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | B | 76 | 2025-06-18+ | v2 |
Parameter relationships underdocumented. seller_plan and seller_level interact (level overrides plan), but this dependency is only mentioned in seller_level description. Both descriptions should cross-reference the relationship.