Paid deterministic CSV cleanup and source-HTML checks. x402 USDC on Base.
Scored across 2 tools
The two tools operate on completely different inputs (CSV files vs. a public HTTPS page), so there is essentially no risk of an agent selecting the wrong one. The only friction is that both descriptions are dominated by identical x402 payment boilerplate, which makes the real distinction take longer to spot.
Both names are snake_case and follow the same resource_qualifier_action shape (csv_import_cleanup, website_html_check), so they read predictably side by side. The pattern is slightly awkward (a qualifier inserted between resource and verb) but it is applied consistently.
Two tools is thin for a server named 'agent-services', which implies a broader utility surface. Each individual tool is well-scoped and non-trivial, so this is borderline rather than an extreme mismatch.
Each tool is explicitly a narrow, single-shot operation that disclaims enrichment, crawling, rendering and fuzzy matching, and there are no companion capabilities (payment/status check, larger-file handling, follow-up operations) to build a workflow. The surface is self-contained but leaves obvious dead ends for anything beyond the two exact use cases.
Costs 25 USDC on Base per paid task; zero model calls. Map columns, trim selected fields and remove exact duplicate output rows for at most 500 rows, 20 columns and 64 KiB UTF-8 CSV. Returns cleaned CSV, quarantined exceptions and an audit. No fuzzy matching or enrichment. Call without payment to obtain x402 requirements; retry with the signed PaymentPayload in params._meta["x402/payment"]. A payment-capable client and authorized buyer wallet are required.
| Name | Required | Description | Default |
|---|---|---|---|
| csv | Yes | At most 64 KiB of UTF-8 CSV. | |
| trim | No | ||
| columns | Yes | ||
| mapping | Yes | ||
| required | No | ||
| unique_key | No |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Annotations already declare non-readonly, non-destructive, non-idempotent, closed-world, so the safety profile is covered. The description adds genuine context beyond that: cost of 25 USDC on Base, zero model calls, hard input bounds, return artifacts, and the payment/wallet authorization prerequisite. It does not say what happens on partial failure of the audit vs. quarantine, but the disclosure is strong.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
Three dense sentences with the cost and payment model front-loaded, then semantics, then constraints. Nearly every clause carries information. Slight congestion from packing pricing, limits, and payment protocol into one paragraph.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
No output schema exists, and the description compensates by naming the three return artifacts (cleaned CSV, quarantined exceptions, audit). Combined with cost, bounds, and the x402 dance, an agent has what it needs to invoke correctly; only the undocumented optional parameters (required, unique_key) leave a gap.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Schema description coverage is only 17%, so the description must compensate. It adds useful constraints (64 KiB CSV, max 20 columns, 500 rows) and implies the trim and duplicate-removal behaviors, but never explains the `required` array, the `unique_key` nullable parameter, or the exact semantics of `mapping` beyond 'map columns'. Partial compensation, not full.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
States specific verbs and resource: map columns, trim fields, remove exact duplicate output rows, returning cleaned CSV plus quarantined exceptions and an audit. It also bounds the operation (500 rows, 20 columns, 64 KiB) and explicitly excludes fuzzy matching/enrichment, which no sibling does, so it is unambiguous against website_html_check.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
Gives a clear procedural path: call without payment to obtain x402 requirements, then retry with the signed PaymentPayload in params._meta["x402/payment"]. It also names the exclusion (no fuzzy matching or enrichment) so the agent knows when this is the wrong tool. Lacking only a direct statement of alternatives/limits on retry behavior.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Costs 25 USDC on Base per paid task; zero model calls. Inspect one authorized public HTTPS page, at most 512 KiB source HTML: HTTP status, title, metadata, headings and link inventory. No JavaScript rendering, crawling, repairs or security/compliance assurance. Call without payment to obtain x402 requirements; retry with the signed PaymentPayload in params._meta["x402/payment"]. A payment-capable client and authorized buyer wallet are required.
| Name | Required | Description | Default |
|---|---|---|---|
| pageUrl | Yes | ||
| permission | Yes | ||
| scopeConsent | Yes |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Beyond the annotations it discloses price (25 USDC on Base), that there are zero model calls, the 512 KiB cap, and the authorization requirement. The readOnlyHint=false annotation is consistent with the description because a paid task moves funds; no contradiction. Cost, auth, and limits are exactly the context annotations cannot express.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
Dense but efficient: cost is front-loaded, then scope, then limits, then the payment protocol. Multiple clauses are packed into few sentences, with no filler, though the result reads as a run-on that could be split for scanability.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
With no output schema, the description usefully enumerates what comes back (HTTP status, title, metadata, headings, links) and documents the two-step payment flow and page limits. Only the semantics of the two consent parameters remain unexplained.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Schema description coverage is 0% and all three parameters (pageUrl, permission, scopeConsent) carry no prose in the schema. The description implies authorization and specifies the x402 payment field, but never explains what permission and scopeConsent mean or that they must be literally true. It compensates only partially.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
States a specific verb and resource ('Inspect one authorized public HTTPS page') plus the exact scope (at most 512 KiB source HTML) and the returned fields (status, title, metadata, headings, link inventory). Nothing about it could be confused with the unrelated sibling csv_import_cleanup.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
Gives explicit exclusions ('no JavaScript rendering, crawling, repairs or security/compliance assurance') and an explicit call protocol: first call without payment to obtain x402 requirements, then retry with the signed PaymentPayload in params._meta['x402/payment']. Prerequisites (payment-capable client, authorized buyer wallet) are named.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
csv_import_cleanupwebsite_html_checkCSV cleanup: dates, emails, phones, dedupe. x402 $0.005 USDC, Base. As is, no warranty.
CSV cleanup: dates, emails, phones, dedupe. x402 $0.005 USDC, Base. As is, no warranty.
Pay-per-call page reader, quote check, link check and x402 seller audit. USDC on Base via x402.
Crypto/DeFi, stock and e-commerce datasets plus pay-per-call web scraping, paid with x402 USDC.
Add one secure layer between your agents and this server.