Overview

SmartGate is an MCP-native algorithm gateway. It sits between your AI agents and the open web, compressing, deduplicating, and hard-capping tool traffic — before runaway agent loops become a line-item surprise.

SmartGate is not an LLM proxy. You keep your own model provider (Cursor, Claude, OpenClaw, etc.) and connect SmartGate as an MCP tool server.

What you can do

• Connect any MCP client that supports Streamable HTTP (MCP 2025-03-26) and Bearer authentication — see MCP Endpoint
• Route agent fetches, searches, compression, dedup, budget checks, memory, and pipelines through one gateway
• Monitor usage, savings estimates, and audit trails in the Dashboard

MCP tools (7)

Tool descriptions and schemas are returned live from the MCP server on tools/list.

Plans at a glance

See Pricing for current pricing. Paid plans bill through Paddle (Merchant of Record).

Get started

1. Quick Start — create an account and API key
2. Connect Guide — copy MCP config from the Dashboard
3. MCP Endpoint — URLs, auth, and audit headers
4. Security — key hygiene and data retention
5. Billing — plans and Paddle subscriptions
6. Teams & Invitations — roles and invite links
7. Tools Reference — all seven smart_* tools
8. Activity Logs — audit, tasks, and usage reports
9. Webhooks (Preview) — register endpoints (delivery coming soon)

Explore further

• Industries — governed AI traffic guidance by topic
• Integrations — connecting SmartGate to the tools you already use
• Features — Token Control, Traffic Customs, Audit & Compliance

Need help? Email support@smartgate.network.

Intelligence layer, not another model product

SmartGate is the intelligence layer between AI and the world. In product language that means five capabilities—Research, Context, Memory, Control, and Pipeline—powered by seven algorithm primitives (smart_search, smart_fetch, smart_dedup, smart_context_gate, smart_memory, smart_budget_guard, smart_pipe). Marketing sometimes lists the tools first; the lasting framing is the capability map: humans get Web scenes without needing to understand MCP, and builders get one Streamable HTTP MCP/API surface so they do not rebuild research infrastructure for every agent.

What SmartGate is not: an LLM proxy, ChatGPT alternative, AI search engine, or model router. You keep Cursor, Claude, OpenClaw, or your own runtime as the host that reasons and chats. SmartGate shapes tool traffic—fetch, search, compress, dedup, budget, memory, and pipes—before that traffic becomes billable model input. Est. USD saved in the dashboard is a gateway counterfactual for comparing tool paths, not a restatement of your vendor LLM invoice.

How the pieces fit on day one

Start with Quick Start to mint a key, then Connect to paste Streamable HTTP JSON into your host. Use Tools Reference when you need parameter-level detail, MCP Endpoint for auth and headers, and Activity Logs when FinOps or security asks what ran. Billing and Teams cover plans and roles; Security covers key hygiene and retention; Webhooks documents the preview surface for outbound endpoints.

Feature pillars on the marketing site map to the same runtime: Token Control for compress, dedup, and budget guard; Traffic Customs for search, fetch, and pipe; Audit and Compliance for team-scoped records and export. Pricing is documented on /pricing with fair-use rate limits on Free/Playground and HTTP 429 soft-fails when bursts exceed entitlements.

Operational expectations

• Single origin: POST /api/mcp for MCP; REST audit and dashboard APIs for humans and automation.
• Team isolation: audit queries and memory are scoped by eam_id.
• Fair-use on Free: MCP RPM per key, team ceilings, and Playground RPM/daily caps are entitlements—not silent drops.
• No invented page text: smart_fetch returns Markdown-friendly content from the URL you asked for; it does not fabricate citations.
• Support: support@smartgate.network for billing, technical, and partnership questions.

Practical adoption notes

Treat SmartGate as shared infrastructure: one gateway URL, team-scoped keys, and audit on by default. Prefer pipe templates for repeatable research paths so Activity Logs show a coherent chain instead of ad-hoc tool spam. Keep host-model prompts focused on decisions; let Research and Context primitives prepare evidence. When FinOps asks for savings, export Usage Reports alongside the gateway benchmark explanation—never present Est. USD saved as a vendor LLM invoice. When security asks for evidence, export an audit window and show retention tied to the current plan catalog.

If you are migrating from scripts, replace raw HTTP fetch helpers first, then add dedup and compression, then budget guard. If you are migrating from a model gateway mindset, stop expecting chat completions on /api/mcp and wire MCP hosts instead. If you are writing human-facing product copy, lead with Smart Research / Smart Decision outcomes; keep MCP JSON on builder paths. These habits keep the site and the product aligned with the Positioning Bible: intelligence layer between AI and the world, not an LLM proxy.

Related reading

• Overview for the capability map and plan table
• Connect Guide for host JSON and troubleshooting
• Tools Reference for when to search, fetch, compress, or pipe
• Activity Logs for audit, trace, and export
• Pricing for fair-use limits, Gainshare, and plan ceilings
• Token Control, Traffic Customs, and Audit and Compliance for pillar narratives aligned with this docs set

Keep English-first public docs consistent with the Positioning Bible: SmartGate is an MCP-native algorithmic gateway and intelligence layer—not a chatbot, ChatGPT alternative, AI search engine, LLM proxy, or model router. Humans research and decide through Web experiences; agents connect through MCP/API; both share the same runtime primitives and team-scoped governance.