com.gatecoreaiPublished by gatecoreai.comOfficial publisherv1.1.0Updated Oct 25, 2026
Procure governed AI capabilities: machine-readable price, scope, trust, gateway-delegation terms.
A remote MCP server for GateCore's governed AI capability marketplace, letting an assistant discover listings, read contract terms, and request procurement.
Web executable via Streamable HTTP. Remote servers run from the web runtime once configured in a workspace.
Paste this prompt into Claude Code, Codex, Cursor or another agent — or into SourceWeft chat. The agent reads this server's install guide, shows you its publisher, endpoint and sign-in, and adds it to your client once you agree. Keys never go into a command.
Read https://sourceweft.com/mcp/com.gatecoreai%2Fmarketplace/install.md and connect the MCP server it describes to my client. Before changing any config, show me its publisher, endpoint and sign-in method and wait for my OK. Never put an API key in a command or in a file I would commit; ask me where it should come from.Add this to your client's mcpServers config.
{
"mcpServers": {
"gatecore-marketplace": {
"type": "http",
"url": "https://mcp.gatecoreai.com/mcp"
}
}
}GateCore is the governed transaction layer for AI agents. Every agent request is identity-verified, policy-checked, priced, settled, and receipted on a tamper-evident ledger before it touches tools, data, or money.
This repository is GateCore's public integration surface: the MCP server manifest, a
quickstart, runnable client examples, and the agent-facing capability file. It is the
canonical reference for the com.gatecoreai/marketplace entry in the official MCP registry.
Point any MCP client at https://mcp.gatecoreai.com/mcp and, with no key and no account:
discover_listings returns published marketplace
listings with machine-readable price, required scopes, and minimum trust terms attached.get_listing returns one listing and its
machine-enforceable contract: price, required scopes, minimum trust, target, and data
classification. Your agent can evaluate the terms programmatically instead of parsing
marketing copy.With a GateCore MCP access key (gcmk_ prefix), an agent can also:
procure returns PROCURE, REVIEW, or DENY.
An eligible decision carries gateway delegation metadata, and execution happens through
GateCore's signed gateway API.list_procurements returns the procurement decisions
visible to the credential's tenant.submit_lead delivers a qualified consumer lead to a
listing that accepts them.QUICKSTART.md has the exact handshake and copy-pasteable calls. examples/ has short, dependency-light clients for discovery and listing fetch.
An agent that wants to buy something on the open web has to read a human sales page, guess at the price, guess at the scope of what it is allowed to do with the result, and take the seller's word for quality. GateCore replaces all three guesses with structured fields on the listing:
Because those terms are attached to the listing rather than to a sales page, the decision to transact is one an agent can make on its own and a human can audit afterward.
Every governed decision produces a signed receipt that binds the request, the policy
decision, and the settlement together. Receipts are Ed25519-signed and independently
verifiable: GateCore publishes the receipt signing keys at
https://api.gatecoreai.com/v1/receipts/keys, and a client-side verifier at
gatecoreai.com/verify recomputes the canonical
bytes and checks the signature entirely in the browser. Verification requires no account
and no call back to GateCore, so a receipt stands as evidence to a counterparty who does
not trust us.
That is the point of the design. The decision is bound to the evidence, so "what did the agent do, under whose authority, at what price, and on what terms" has one answer that survives the session.
GateCore is listed on the official Model Context Protocol registry as
com.gatecoreai/marketplace. server.json in this repository is that
manifest.
Discovery is open. Procurement, procurement history, and lead submission require a GateCore-issued MCP access key so that agent identity and tenant scope come from the credential rather than from tool-call arguments. Get one self-serve, free, at app.gatecoreai.com/developer/portal. For anything else, contact [email protected].
Source: README.md at commit 7056a2f
Discover Marketplace Listings
Discover governed GateCore marketplace capabilities with machine-readable pricing, required scopes, and minimum trust terms. Results preserve the marketplace's organic order.
tagsquerymin_trustmax_price_centsGet Listing Details
Fetch one governed marketplace listing and its machine-enforceable contract terms: price, required scopes, minimum trust, target, and data classification.
listing_idrequiredRequest Governed Procurement
Request governed procurement for a listing. In public full mode, identity comes from the MCP access key, trust comes from GateCore's baseline, and the result is a PROCURE, REVIEW, or DENY decision with gateway delegation when eligible.
scopeslisting_idrequiredrequest_idprocurement_idmax_price_centsrequester_trustrequiredrequester_agent_idrequiredAuthorize Governed Access
Authorize signed access to a governed listing. Purpose is forwarded unchanged when present; policy at the gateway decides whether the declared purpose is allowed.
purposelisting_idrequiredmax_price_centsComplete Governed Access
Hash seller output locally, attest completion with the buyer agent key, and return the terminal signed receipt. Seller output itself is never forwarded or returned.
ticketrequiredoutcomerequiredseller_outputrequiredList Procurement Decisions
List procurement decisions for the credential tenant in public modes.
Takes no parameters.
Request Listing Access
Submit an idempotent listing-access request for the credential's tenant and agent.
agent_idlisting_idrequiredSubmit Consumer Lead
Submit a consented consumer lead to a supported published listing.
leadrequiredlisting_idrequired100% of the applicable checks met
How complete this listing is, computed from the registry entry, the endpoint and public sources. Not a rating of the server and not a statement that it is safe or working today: reachability is a dated check, and no installs or reviews go into the score.