Agent-native MCP for governed commerce, x402 payments, paid capabilities, and verifiable receipts.
Scored across 28 tools
Several free discovery/read-only tools overlap in surface (capability.search, catalog.list, offers.list, provider.describe, marketplace.describe), and multiple paid audit tools target related metadata and endpoint checks. The descriptions explicitly clarify boundaries with 'Do NOT use' notes, but the agent still faces meaningful misselection risk across 28 tools.
Most tools follow a predictable dotted namespace.action pattern (capability.get, catalog.list, commerce.quote, x402.compatibility_audit). A few names use noun-phrase actions or multiword underscores, but the overall convention is readable and largely consistent.
With 28 tools, the server is above the 25-tool threshold and carries heavy cognitive load. The platform spans marketplace, provider, audit, and utility functions, but several tools could likely be consolidated without losing core capability.
The surface covers discovery, quoting, preflight, preview, and audits, but it lacks a direct purchase/execution or entitlement/receipt tool. The commerce.quote description references capability.purchase, yet no such tool is present, creating a significant gap for end-to-end marketplace workflows.
PAID $0.10 USDC on Base via x402. Creates deterministic SHA-256 entries and a root manifest hash for supplied text artifacts. Calling this MCP tool does not spend funds; it returns the exact PAYMENT-REQUIRED challenge and canonical paid execution endpoint.
Inputs: artifacts (required). Returns: asset, error, input, price, +17 more
FREE. Use name for legacy deterministic capability metadata or capabilityId for public marketplace metadata, rights and provenance.
Inputs: name, capabilityId. Returns: object
FREE MARKETPLACE FILTERED DISCOVERY ONLY. Search active public marketplace capabilities when one or more filters are known: text, category, provider, delivery type, or price. Metadata is free; purchased execution and deliverables are not. Do NOT use for unfiltered browsing; use catalog.list. Do NOT use for CrossingKey digital/service offers; use offers.list. For one exact capabilityId use capability.get.
Inputs: limit, query, offset, category, +4 more. Returns: object
FREE MARKETPLACE UNFILTERED DISCOVERY ONLY. Page through active public marketplace capabilities when no search filters are needed. Metadata is free; execution, entitlement, delivery, and purchased output remain gated. Do NOT use for filtered discovery; use capability.search. Do NOT use for CrossingKey digital/service offers; use offers.list. For one exact capabilityId use capability.get.
Inputs: limit, offset. Returns: object
FREE MARKETPLACE PAYMENT QUOTE ONLY. After selecting an active marketplace capabilityId, return authoritative gross price, CrossingKey fee, provider share, and x402 v1/v2 payment requirements. A quote grants no entitlement and performs no execution. Do NOT use for first-party paid capabilities; use cost.estimate (first-party x402 capabilities have fixed catalog prices and no quote stage). Do NOT use for generic price lookup; use cost.estimate. Paid marketplace execution requires capability.purchase with authorization and verified payment.
Inputs: capabilityId (required). Returns: object
FREE read-only cost estimate. This tool never creates checkout, reserves funds, or contacts a facilitator.
Inputs: item_id (required). Returns: cost, found, item_id, cost_state, +2 more
FREE WRITE. Private creator intake for assets needing a capability model. No ownership transfer and AI training defaults false.
Inputs: summary (required), provider (required). Returns: object
FREE read-only legacy credit compatibility status. New prepaid-credit sales are disabled in v4; historical paid credits remain redeemable through a separate compatibility path.
Inputs: no named parameters. Returns: free, packs, payment_started
FREE read-only execution preflight. Returns READY_FOR_HUMAN_AUTHORIZATION only; it never authorizes, purchases, settles, executes, or creates records.
Inputs: item_id (required). Returns: state, checks, reason, item_id, +6 more
PAID $0.01 USDC on Base via x402. Computes a SHA-256 digest and UTF-8 byte count for supplied text. Calling this MCP tool returns a payment challenge and does not autonomously spend funds.
Inputs: text (required). Returns: asset, error, input, price, +17 more
PAID $0.01 USDC on Base via x402. Validates JSON and repairs a limited set of formatting errors. May return supplied content. Calling this MCP tool returns a payment challenge and does not autonomously spend funds.
Inputs: text (required). Returns: asset, error, input, price, +17 more
PAID $2.00 USDC on Base via x402. Checks machine-commerce discovery and health surfaces. Calling this MCP tool does not spend funds; it returns the exact PAYMENT-REQUIRED challenge and canonical paid execution endpoint.
Inputs: url (required). Returns: asset, error, input, price, +17 more
FREE, read-only marketplace overview. Use this first for marketplace-specific capability commerce, fee policy, creator rights, payment rails, and active catalog counts; use provider.describe for the broader CrossingKey provider and legacy offer surface.
Inputs: no named parameters. Returns: object
PAID $9.00 USDC on Base via x402. Performs bounded live MCP discovery, health, transport, schema, and receiver-only checks against a public endpoint. No payment is attempted by the audit.
Inputs: url (required). Returns: asset, error, input, price, +17 more
PAID $1.00 USDC on Base via x402. Scores supplied MCP tool schemas for discoverability, descriptions, and bounded input contracts. Calling this MCP tool does not spend funds; it returns the exact PAYMENT-REQUIRED challenge and canonical paid execution endpoint.
Inputs: tools (required). Returns: asset, error, input, price, +17 more
FREE read-only public offer and capability discovery. This tool never starts payment or execution.
Inputs: limit, query, offset. Returns: free, total, offers, next_offset, +1 more
PAID $1.00 USDC on Base via x402. Evaluates structural quality and required fields in an OpenAPI document. Calling this MCP tool does not spend funds; it returns the exact PAYMENT-REQUIRED challenge and canonical paid execution endpoint.
Inputs: spec (required). Returns: asset, error, input, price, +17 more
PAID $0.01 USDC on Base via x402. Detects selected personal-data patterns and returns classifications and positions, not matched values. Heuristic, not comprehensive. Calling this MCP tool returns a payment challenge and does not autonomously spend funds.
Inputs: text (required). Returns: asset, error, input, price, +17 more
PAID $0.01 USDC on Base via x402. Scores selected prompt structure signals using deterministic heuristics. Calling this MCP tool returns a payment challenge and does not autonomously spend funds.
Inputs: text (required). Returns: asset, error, input, price, +17 more
FREE public application intake. Always pending, never creates credentials or active capabilities.
Inputs: slug (required), contact (required), description (required), displayName (required), +1 more. Returns: object
FREE read-only provider and commerce policy discovery. Use this before selecting an offer or capability.
Inputs: no named parameters. Returns: manifest, provider, sequence, stopBeforePayment
FREE. Public active provider metadata. Pending applications require owner or administrator authorization.
Inputs: providerId (required). Returns: object
FREE read-only requirements check. This tool never authorizes or starts payment or execution.
Inputs: item_id (required). Returns: type, found, item_id, requirements, +2 more
FREE read-only result-shape preview. This tool never executes a capability or creates an entitlement or receipt.
Inputs: item_id (required). Returns: found, item_id, preview_state, result_preview, +1 more
PAID $0.01 USDC on Base via x402. Detects selected credential patterns and returns classifications and positions, not matched values. Heuristic, not comprehensive. Calling this MCP tool returns a payment challenge and does not autonomously spend funds.
Inputs: text (required). Returns: asset, error, input, price, +17 more
PAID $1.00 USDC on Base via x402. Audits an x402 endpoint for v1/v2 compatibility defects. Calling this MCP tool does not spend funds; it returns the exact PAYMENT-REQUIRED challenge and canonical paid execution endpoint.
Inputs: url (required). Returns: asset, error, input, price, +17 more
PAID $19.00 USDC on Base via x402. Performs a bounded no-payment purchase-challenge audit for a selected capability, validating status, requirements, destination, amount, and receiver-only controls.
Inputs: url (required), capability (required). Returns: asset, error, input, price, +17 more
PAID $0.10 USDC on Base via x402. Validates and normalizes one bounded XKEY intake payload. Calling this MCP tool returns the exact PAYMENT-REQUIRED challenge and does not spend funds.
Inputs: raw_intake (required), idempotency_key (required). Returns: asset, error, input, price, +17 more
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
Governed agent execution: x402 payments, budgets, receipts, verification, and audit.
Workflow diagnostics, capability routing, and x402 settlement for MCP-compatible agents.
Authorize and prove autonomous x402/AP2 agent payments with cryptographic receipts.
Governed MCP: agent audit, provenance, deterministic checks, and receipt-backed FragGate execution.
Add one secure layer between your agents and this server.
