{
 "id": "https://redfinchadvisory.com/lexicon/agent-gateway",
 "slug": "agent-gateway",
 "prefLabel": "Agent gateway",
 "altLabel": [
  "Tool gateway",
  "MCP gateway",
  "Tool-access gateway"
 ],
 "hiddenLabel": [],
 "definition": "A control point in front of everything an agent can act on (tools, APIs and MCP servers) that governs what it can reach, enforces least privilege and logs its actions.",
 "scopeNote": "Used three ways today: the outbound control point for agent actions; an MCP gateway in front of your tool servers; and a registry where agents get their own identities and permissions. The control point and the registry are two views of one job: where rules are enforced vs where they are stored. Once an agent can act, the risk is a real side effect (a write, a payment, a deletion), not just a bad answer.",
 "tier": "reference",
 "group": "Connections and systems",
 "broader": [],
 "narrower": [],
 "related": [
  {
   "id": "https://redfinchadvisory.com/lexicon/llm-gateway",
   "type": "confused"
  },
  {
   "id": "https://redfinchadvisory.com/lexicon/stack",
   "type": "related"
  }
 ],
 "links": [
  {
   "id": "https://redfinchadvisory.com/lexicon/agent-identity",
   "label": "reads",
   "direction": "out"
  },
  {
   "id": "https://redfinchadvisory.com/lexicon/least-privilege-for-ai",
   "label": "enforced by",
   "direction": "in"
  },
  {
   "id": "https://redfinchadvisory.com/lexicon/tool",
   "label": "is governed by",
   "direction": "in"
  },
  {
   "id": "https://redfinchadvisory.com/lexicon/mcp-server",
   "label": "sits in front of",
   "direction": "out"
  }
 ],
 "modified": "2026-10-21",
 "markdown": "https://redfinchadvisory.com/lexicon/agent-gateway.md",
 "inScheme": {
  "id": "https://redfinchadvisory.com/lexicon",
  "version": "1.0",
  "currentAsOf": "2026-10-21"
 }
}