Skip to main content
Glama
Sign In
deenesjakoruzh
AI Gateway MCP

AI Gateway MCP

by CSOAI-ORG
OverviewSchemaSourceRelated ServersDiscussionsAdmin
Python

MCP Scorecard: 86/100

Ai Gateway MCP

MEOK AI Labs GSPC License PyPI

Ai Gateway MCP ā€” automation tooling for ai gateway

Ai Gateway MCP ā€” automation tooling for ai gateway. MIT.

šŸš€ Quick Start

# Install via pip
pip install ai_gateway_mcp

# Or install via Smithery
npx -y @smithery/cli@latest install ai-gateway-mcp --client claude

Related MCP server: AI Ops MCP

āœØ Features

  • MCP protocol compliant

  • Easy installation

  • Well-documented API

  • Production-ready

  • Active maintenance

šŸ“– Documentation

šŸ›”ļø Compliance

This MCP server is built with EU AI Act compliance built-in:

  • āœ… Article 9 ā€” Risk Management System

  • āœ… Article 13 ā€” Transparency & Instructions for Use

  • āœ… Article 15 ā€” Bias Detection & Testing

  • āœ… Article 26 ā€” FRIA Support (where applicable)

  • āœ… Article 50 ā€” AI Content Watermarking (where applicable)

Need help getting compliant? Book a free 15-min diagnostic ā†’

šŸ¢ Enterprise

Need custom development, SLA guarantees, or white-label deployment?

  • Pro: $99/mo ā€” Full MCP suite + EU AI Act tracking

  • Enterprise: $499/mo ā€” Custom dev + SLA + Dedicated support

View Pricing ā†’ | Contact Sales ā†’

šŸ¤ Part of the MEOK Ecosystem

This server is part of the MEOK AI Labs ecosystem ā€” 300+ MCP servers for sovereign AI governance.

Domain

Purpose

councilof.ai

EU AI Act compliance marketplace

safetyof.ai

AI safety & monitoring

meok.ai

Sovereign AI platform

cobolbridge.ai

Legacy modernization

šŸ“œ License

MIT Ā© CSOAI-ORG

Build something that touches users? You need compliance. MEOK ships 38 governance MCPs that drop in alongside this tool ā€” EU AI Act, DORA, NIS2, CRA, GDPR, ISO 42001, FDA SaMD, MDR, Basel, MiFID II, MiCA, COPPA, and more.

# One-shot install of the governance pack
npx meok-setup --pack governance

Free tier: 10 calls/day per MCP. Pro tier (Ā£79/mo): unlimited + cryptographically signed compliance attestations your auditor verifies independently.

ā†’ Full catalogue: councilof.ai/catalogue ā†’ MEOK AI Labs: meok.ai

Protocol coverage + Universal PAYG

This MCP is part of MEOK's 47-MCP fleet that bridges every active agent-interop protocol and 30+ regulatory frameworks. See the full coverage matrix at meok.ai/protocols.

Agent interop protocols supported (8 live):

  • āœ… MCP (Anthropic) ā€” native

  • āœ… A2A (Google + Linux Foundation, absorbed IBM ACP Sept 2025)

  • āœ… IBM ACP ā€” covered via A2A merge

  • ā— Stripe ACP (Agentic Commerce Protocol) ā€” Q3 bridge via agent-commerce-protocol-mcp

  • ā— AP2 (Google Agent Payments) ā€” partial via agent-commerce-payments-mcp

  • ā— x402 (Coinbase HTTP 402) ā€” partial via api.meok.ai gateway

  • ā†’ OASF / AGNTCY (Cisco Outshift + Linux Foundation) ā€” Q3 bridge

  • šŸ‘ ANP (Cisco Agent Network) ā€” watch-list

Pricing options:

Option

Price

Best for

Self-host (this MCP)

Ā£0 ā€” MIT

Devs

This MCP Starter

Ā£29/mo

One-MCP teams

This MCP Pro

Ā£79/mo

Production + 24h SLA

Universal PAYG

Ā£29/mo + Ā£0.0002/call

Spiky usage across many MCPs

Substrate bundle (this category)

Ā£99-Ā£499/mo

A whole pack

MEOK Universe

Ā£1,499/mo

All 47 MCPs, 500K calls

Each tier above the free self-host adds HMAC-signed attestations verifiable at verify.meok.ai. Linux Foundation governance on the A2A spine means EU regulated buyers can deploy without vendor-lock-in objections.

šŸ’ø Try MEOK in 30 seconds ā€” instant buy ladder

Tier

Price

What you get

Stripe

Smoke test

Ā£1

Signed sample MCP-Hardening report + Article 50 PDF

https://buy.stripe.com/aFa7sNcgAdQS0ZT1Uc8k91t

Quick Kit

Ā£9

EU AI Act Article 50 implementation guide (C2PA + EU-Icon)

https://buy.stripe.com/aFa7sNcgAdQS0ZT1Uc8k91t

Founder Call

Ā£29

30-min 1-on-1 with the founder

https://buy.stripe.com/aFa7sNcgAdQS0ZT1Uc8k91t

Refundable. UK Stripe ā€” VAT-clean. Builds on the 81-MCP MEOK fleet. Verify any signed report at https://meok.ai/verify.

Configuration

Add to your claude_desktop_config.json (Claude Desktop) or your MCP client config:

{
  "mcpServers": {
    "ai-gateway-mcp": {
      "command": "uvx",
      "args": ["ai-gateway-mcp"]
    }
  }
}

Or: pip install ai-gateway-mcp then run the ai-gateway-mcp command (stdio transport).

Examples

Once configured, ask your assistant, for example:

  • "Use route_request to ā€¦"

  • "Use list_models to ā€¦"

  • "Use cost_estimator to ā€¦"

Available Tools

4 tools
cost_estimatorA

Estimate cost for a specific request across models.

Behavior: This tool is read-only and stateless ā€” it produces analysis output without modifying any external systems, databases, or files. Safe to call repeatedly with identical inputs (idempotent). Free tier: 10/day rate limit. Pro tier: unlimited. No authentication required for basic usage.

When to use: Use this tool when you need structured analysis or classification of inputs against established frameworks or standards.

When NOT to use: Not suitable for real-time production decision-making without human review of results.

Args: prompt_tokens (int): The prompt tokens to analyze or process. completion_tokens (int): The completion tokens to analyze or process. model (str): The model to analyze or process. api_key (str): The api key to analyze or process.

Behavioral Transparency: - Side Effects: This tool is read-only and produces no side effects. It does not modify any external state, databases, or files. All output is computed in-memory and returned directly to the caller. - Authentication: No authentication required for basic usage. Pro/Enterprise tiers require a valid MEOK API key passed via the MEOK_API_KEY environment variable. - Rate Limits: Free tier: 10 calls/day. Pro tier: unlimited. Rate limit headers are included in responses (X-RateLimit-Remaining, X-RateLimit-Reset). - Error Handling: Returns structured error objects with 'error' key on failure. Never raises unhandled exceptions. Invalid inputs return descriptive validation errors. - Idempotency: Fully idempotent ā€” calling with the same inputs always produces the same output. Safe to retry on timeout or transient failure. - Data Privacy: No input data is stored, logged, or transmitted to external services. All processing happens locally within the MCP server process.

Inputs: prompt_tokens (required), completion_tokens (required), model, api_key. Returns: result

get_gateway_statsA

Get gateway usage statistics.

Behavior: This tool is read-only and stateless ā€” it produces analysis output without modifying any external systems, databases, or files. Safe to call repeatedly with identical inputs (idempotent). Free tier: 10/day rate limit. Pro tier: unlimited. No authentication required for basic usage.

When to use: Use this tool when you need structured analysis or classification of inputs against established frameworks or standards.

When NOT to use: Not suitable for real-time production decision-making without human review of results.

Args: api_key (str): The api key to analyze or process.

Behavioral Transparency: - Side Effects: This tool is read-only and produces no side effects. It does not modify any external state, databases, or files. All output is computed in-memory and returned directly to the caller. - Authentication: No authentication required for basic usage. Pro/Enterprise tiers require a valid MEOK API key passed via the MEOK_API_KEY environment variable. - Rate Limits: Free tier: 10 calls/day. Pro tier: unlimited. Rate limit headers are included in responses (X-RateLimit-Remaining, X-RateLimit-Reset). - Error Handling: Returns structured error objects with 'error' key on failure. Never raises unhandled exceptions. Invalid inputs return descriptive validation errors. - Idempotency: Fully idempotent ā€” calling with the same inputs always produces the same output. Safe to retry on timeout or transient failure. - Data Privacy: No input data is stored, logged, or transmitted to external services. All processing happens locally within the MCP server process.

Inputs: api_key. Returns: result

list_modelsC

List all available models with capabilities and pricing.

Behavior: This tool is read-only and stateless ā€” it produces analysis output without modifying any external systems, databases, or files. Safe to call repeatedly with identical inputs (idempotent). Free tier: 10/day rate limit. Pro tier: unlimited. No authentication required for basic usage.

When to use: Use this tool when you need structured analysis or classification of inputs against established frameworks or standards.

When NOT to use: Not suitable for real-time production decision-making without human review of results.

Args: filter_provider (str): The filter provider to analyze or process. filter_compliance (str): The filter compliance to analyze or process. api_key (str): The api key to analyze or process.

Behavioral Transparency: - Side Effects: This tool is read-only and produces no side effects. It does not modify any external state, databases, or files. All output is computed in-memory and returned directly to the caller. - Authentication: No authentication required for basic usage. Pro/Enterprise tiers require a valid MEOK API key passed via the MEOK_API_KEY environment variable. - Rate Limits: Free tier: 10 calls/day. Pro tier: unlimited. Rate limit headers are included in responses (X-RateLimit-Remaining, X-RateLimit-Reset). - Error Handling: Returns structured error objects with 'error' key on failure. Never raises unhandled exceptions. Invalid inputs return descriptive validation errors. - Idempotency: Fully idempotent ā€” calling with the same inputs always produces the same output. Safe to retry on timeout or transient failure. - Data Privacy: No input data is stored, logged, or transmitted to external services. All processing happens locally within the MCP server process.

Inputs: api_key, filter_provider, filter_compliance. Returns: result

route_requestA

Route an AI request to the optimal model based on task, cost, speed, and compliance requirements.

Behavior: This tool is read-only and stateless ā€” it produces analysis output without modifying any external systems, databases, or files. Safe to call repeatedly with identical inputs (idempotent). Free tier: 10/day rate limit. Pro tier: unlimited. No authentication required for basic usage.

When to use: Use this tool when you need structured analysis or classification of inputs against established frameworks or standards.

When NOT to use: Not suitable for real-time production decision-making without human review of results.

Args: task (str): The task to analyze or process. priority (str): The priority to analyze or process. max_cost (float): The max cost to analyze or process. require_compliance (str): The require compliance to analyze or process. prefer_local (bool): The prefer local to analyze or process. api_key (str): The api key to analyze or process.

Behavioral Transparency: - Side Effects: This tool is read-only and produces no side effects. It does not modify any external state, databases, or files. All output is computed in-memory and returned directly to the caller. - Authentication: No authentication required for basic usage. Pro/Enterprise tiers require a valid MEOK API key passed via the MEOK_API_KEY environment variable. - Rate Limits: Free tier: 10 calls/day. Pro tier: unlimited. Rate limit headers are included in responses (X-RateLimit-Remaining, X-RateLimit-Reset). - Error Handling: Returns structured error objects with 'error' key on failure. Never raises unhandled exceptions. Invalid inputs return descriptive validation errors. - Idempotency: Fully idempotent ā€” calling with the same inputs always produces the same output. Safe to retry on timeout or transient failure. - Data Privacy: No input data is stored, logged, or transmitted to external services. All processing happens locally within the MCP server process.

Inputs: task (required), api_key, max_cost, priority, +2 more. Returns: result

Tool Schema Changelog

Recent tool additions, removals, and schema changes observed during successful MCP inspections.

Loading tool schema changelogā€¦

    Deploy Server
    Claude Code, inside the Godot 4 editorFlockbay puts an AI agent in the real Godot 4 editor: it writes GDScript, runs your game and fixes what breaks. Free, or bring your own Claude Code or Codex.flockbay.comAd

    TDQS

    A3.9/5.0

    Scored across 4 tools

    Disambiguation5/5

    Each tool has a clearly distinct purpose: cost estimation, gateway stats, model listing, and request routing. No overlap in functionality.

    Naming Consistency5/5

    All tool names follow a consistent verb_noun pattern in snake_case (cost_estimator, get_gateway_stats, list_models, route_request).

    Tool Count5/5

    4 tools are well-scoped for an AI gateway server, covering essential capabilities without bloat or insufficiency.

    Completeness4/5

    The tools cover core gateway operations (cost, stats, model info, routing). Minor gaps like API key management or log retrieval are acceptable for a focused set.

    Maintenance

    ActivityStale
    ResponsivenessNo recent community issues

    Resources

    Related MCP Connectors

    Related MCP Servers