MCPGateway

MCPGateway is a universal MCP aggregation server that exposes a single HTTP/SSE endpoint (/mcp and /sse) to route to multiple upstream MCP servers (STDIO/HTTP/SSE), while applying aggressive token/context optimizations, result filtering/aggregation, sandboxed code execution, and providing a web dashboard for managing backends and tools.

Evaluated Mar 30, 2026 (22d ago)
Repo ↗ Infrastructure mcp gateway routing token-optimization dashboard typescript sse http code-execution observability aggregation
⚙ Agent Friendliness
60
/ 100
Can an agent use this?
🔒 Security
62
/ 100
Is it safe for agents?
⚡ Reliability
36
/ 100
Does it work consistently?

Score Breakdown

⚙ Agent Friendliness

MCP Quality
78
Documentation
70
Error Messages
0
Auth Simplicity
70
Rate Limits
35

🔒 Security

TLS Enforcement
85
Auth Strength
70
Scope Granularity
40
Dep. Hygiene
55
Secret Handling
55

Provides AUTH_MODE=none/api-key/oauth plus mention of JWT; blocks sensitive endpoints when auth is disabled unless ALLOW_INSECURE=1. Also lists helmet/cors usage and includes an audit logging claim. Specific TLS enforcement, rotation policies, and detailed security controls for code-execution are not fully verifiable from provided excerpt.

⚡ Reliability

Uptime/SLA
10
Version Stability
35
Breaking Changes
45
Error Recovery
55
AF Security Reliability

Best When

You need a single MCP endpoint for heterogeneous tool servers plus substantial context/token savings and a management dashboard.

Avoid When

You cannot lock down authentication for sensitive endpoints, or you need a fully specified, strongly contract-tested REST API with comprehensive machine-readable schemas.

Use Cases

  • Expose many MCP tools from multiple upstream servers through one endpoint for AI clients
  • Reduce token/context usage by progressive tool discovery, result filtering, batching, deduplication, delta responses, and auto-summarization
  • Run sandboxed TypeScript/JavaScript operations via gateway code-execution MCP tools
  • Operate a centralized dashboard to add/manage/reconnect MCP backends and toggle tool enablement
  • Use Prometheus and JSON metrics endpoints for gateway observability

Not For

  • High-security, multi-tenant production deployments without careful security configuration and secret management
  • Environments requiring formal, published SLAs or strong guarantees around API stability
  • Use cases where HTTP endpoint access must be avoided (gateway is inherently an HTTP service)
  • Systems that require an official OpenAPI spec or guaranteed stable API contract for the dashboard/code endpoints

Interface

REST API
Yes
GraphQL
No
gRPC
No
MCP Server
Yes
SDK
Yes
Webhooks
No

Authentication

Methods: API key OAuth JWT
OAuth: Yes Scopes: No

README indicates AUTH_MODE supports none/api-key/oauth and API_KEYS or OAUTH_* settings; it also states sensitive endpoints are blocked by default when running with AUTH_MODE=none unless ALLOW_INSECURE=1 is set.

Pricing

Free tier: No
Requires CC: No

Open-source repository; no vendor pricing described in provided content.

Agent Metadata

Pagination
none
Idempotent
False
Retry Guidance
Not documented

Known Gotchas

  • Authentication modes affect access to sensitive endpoints (/dashboard, /dashboard/api/*, /api/code/*, /metrics/json). Agents should not assume unauthenticated access works by default.
  • Code execution endpoint (/api/code/execute) may be restricted/secured in practice; ensure sandbox and permissions are configured appropriately.
  • Token-optimization behaviors (delta responses, session context) may cause surprising outputs if an agent expects raw upstream responses.

Alternatives

Full Evaluation Report

Comprehensive deep-dive: security analysis, reliability audit, agent experience review, cost modeling, competitive positioning, and improvement roadmap for MCPGateway.

AI-powered analysis · PDF + markdown · Delivered within 30 minutes

$99

Package Brief

Quick verdict, integration guide, cost projections, gotchas with workarounds, and alternatives comparison.

Delivered within 10 minutes

$3

Score Monitoring

Get alerted when this package's AF, security, or reliability scores change significantly. Stay ahead of regressions.

Continuous monitoring

$3/mo

Scores are editorial opinions as of 2026-03-30.

8642
Packages Evaluated
17761
Need Evaluation
586
Need Re-evaluation
Community Powered