aid: kong-ai-gateway
name: Kong AI Gateway
description: Kong AI Gateway is the AI-native capability layer built on top of Kong Gateway and managed through Kong Konnect.
  It exposes a normalized, provider-agnostic LLM API across 16+ providers (OpenAI, Anthropic, Azure AI, Amazon Bedrock, Google
  Gemini, Vertex AI, Cohere, Hugging Face, Llama, Mistral, xAI, DashScope, Cerebras, Ollama, Databricks, DeepSeek, vLLM),
  and adds prompt firewalls, PII sanitization, semantic caching, automated RAG injection, token-level observability, per-agent
  cost allocation, MCP traffic governance, and Agent Gateway support for agent-to-agent (A2A) communication. It is profiled
  here as a standalone product surface; the parent provider profile lives at github.com/api-evangelist/kong.
type: Index
image: https://kinlane-productions.s3.amazonaws.com/apis-json/apis-json-logo.jpg
tags:
- AI Gateway
- LLM
- MCP
- A2A
- AI Governance
- Konnect
- Kong
url: https://raw.githubusercontent.com/api-evangelist/kong-ai-gateway/refs/heads/main/apis.yml
created: '2026-05-23'
modified: '2026-05-23'
specificationVersion: '0.19'
apis:
- aid: kong-ai-gateway:kong-ai-gateway
  name: Kong AI Gateway
  description: Kong AI Gateway is the connectivity and governance layer for AI-native applications. Built on Kong Gateway,
    it provides a universal LLM API across 16+ providers, semantic caching, prompt firewalls and PII guardrails, automated
    RAG injection, token-level observability, MCP traffic governance, and Agent Gateway support for agent-to-agent (A2A)
    communication. Traffic flows through configurable Kong routes using the AI Proxy and AI Proxy Advanced plugins.
  humanURL: https://konghq.com/products/kong-ai-gateway
  tags:
  - AI Gateway
  - LLM
  - Plugin
  properties:
  - type: Documentation
    url: https://developer.konghq.com/ai-gateway/
  - type: GettingStarted
    url: https://developer.konghq.com/ai-gateway/get-started/
  - type: APIReference
    url: https://developer.konghq.com/ai-gateway/ai-providers/
  - type: ChangeLog
    url: https://developer.konghq.com/gateway/changelog/
  - type: GitHubRepository
    url: https://github.com/Kong/kong
- aid: kong-ai-gateway:ai-proxy-plugin
  name: AI Proxy Plugin
  description: The AI Proxy plugin transforms and proxies requests to a configured AI provider and model, shielding client
    applications from provider-specific request and response shapes.
  humanURL: https://developer.konghq.com/plugins/ai-proxy/
  tags:
  - Plugin
  - AI Proxy
  - LLM
  properties:
  - type: Documentation
    url: https://developer.konghq.com/plugins/ai-proxy/
- aid: kong-ai-gateway:ai-proxy-advanced-plugin
  name: AI Proxy Advanced Plugin
  description: The AI Proxy Advanced plugin extends AI Proxy with load balancing, weighted distribution, and fallback across
    multiple providers and models simultaneously.
  humanURL: https://developer.konghq.com/plugins/ai-proxy-advanced/
  tags:
  - Plugin
  - Load Balancing
  - LLM
  properties:
  - type: Documentation
    url: https://developer.konghq.com/plugins/ai-proxy-advanced/
- aid: kong-ai-gateway:ai-rate-limiting-advanced-plugin
  name: AI Rate Limiting Advanced Plugin
  description: Token-aware rate limiting tailored for LLM traffic, with per-consumer and per-model budgets rather than just
    request counts.
  humanURL: https://developer.konghq.com/plugins/ai-rate-limiting-advanced/
  tags:
  - Plugin
  - Rate Limiting
  - Token Budget
  properties:
  - type: Documentation
    url: https://developer.konghq.com/plugins/ai-rate-limiting-advanced/
- aid: kong-ai-gateway:ai-prompt-guard-plugin
  name: AI Prompt Guard Plugin
  description: Enforces allow- and deny-lists for prompts and text completions, blocking disallowed content before it reaches
    the model.
  humanURL: https://developer.konghq.com/plugins/ai-prompt-guard/
  tags:
  - Plugin
  - Prompt Firewall
  - Safety
  properties:
  - type: Documentation
    url: https://developer.konghq.com/plugins/ai-prompt-guard/
- aid: kong-ai-gateway:ai-semantic-prompt-guard-plugin
  name: AI Semantic Prompt Guard Plugin
  description: Topic-aware variant of AI Prompt Guard that classifies prompts by meaning and blocks restricted topics regardless
    of phrasing.
  humanURL: https://developer.konghq.com/plugins/ai-semantic-prompt-guard/
  tags:
  - Plugin
  - Prompt Firewall
  - Semantic
  properties:
  - type: Documentation
    url: https://developer.konghq.com/plugins/ai-semantic-prompt-guard/
- aid: kong-ai-gateway:ai-pii-sanitizer-plugin
  name: AI PII Sanitizer Plugin
  description: Detects and redacts personally identifiable information from prompts and responses traversing the gateway.
  humanURL: https://developer.konghq.com/plugins/ai-pii-sanitizer/
  tags:
  - Plugin
  - PII
  - Privacy
  properties:
  - type: Documentation
    url: https://developer.konghq.com/plugins/ai-pii-sanitizer/
- aid: kong-ai-gateway:ai-semantic-cache-plugin
  name: AI Semantic Cache Plugin
  description: Caches LLM responses by prompt similarity so semantically equivalent requests can be served from cache, reducing
    latency and provider spend.
  humanURL: https://developer.konghq.com/plugins/ai-semantic-cache/
  tags:
  - Plugin
  - Caching
  - Semantic
  properties:
  - type: Documentation
    url: https://developer.konghq.com/plugins/ai-semantic-cache/
- aid: kong-ai-gateway:ai-rag-injector-plugin
  name: AI RAG Injector Plugin
  description: Automates retrieval-augmented generation by injecting retrieved context into prompts at the gateway, so application
    code does not need to implement RAG plumbing.
  humanURL: https://developer.konghq.com/plugins/ai-rag-injector/
  tags:
  - Plugin
  - RAG
  - Retrieval
  properties:
  - type: Documentation
    url: https://developer.konghq.com/plugins/ai-rag-injector/
- aid: kong-ai-gateway:ai-prompt-template-plugin
  name: AI Prompt Template Plugin
  description: Provides reusable, fill-in-the-blank prompt templates managed at the gateway layer.
  humanURL: https://developer.konghq.com/plugins/ai-prompt-template/
  tags:
  - Plugin
  - Prompt Engineering
  properties:
  - type: Documentation
    url: https://developer.konghq.com/plugins/ai-prompt-template/
- aid: kong-ai-gateway:ai-prompt-decorator-plugin
  name: AI Prompt Decorator Plugin
  description: Prepends or appends messages to chat history before requests reach the model.
  humanURL: https://developer.konghq.com/plugins/ai-prompt-decorator/
  tags:
  - Plugin
  - Prompt Engineering
  properties:
  - type: Documentation
    url: https://developer.konghq.com/plugins/ai-prompt-decorator/
- aid: kong-ai-gateway:ai-prompt-compressor-plugin
  name: AI Prompt Compressor Plugin
  description: Reduces prompt token count before forwarding to the provider to lower latency and cost.
  humanURL: https://developer.konghq.com/plugins/ai-prompt-compressor/
  tags:
  - Plugin
  - Token Optimization
  properties:
  - type: Documentation
    url: https://developer.konghq.com/plugins/ai-prompt-compressor/
- aid: kong-ai-gateway:ai-azure-content-safety-plugin
  name: AI Azure Content Safety Plugin
  description: Integrates Azure AI Content Safety for content moderation on prompts and responses.
  humanURL: https://developer.konghq.com/plugins/ai-azure-content-safety/
  tags:
  - Plugin
  - Content Safety
  - Azure
  properties:
  - type: Documentation
    url: https://developer.konghq.com/plugins/ai-azure-content-safety/
- aid: kong-ai-gateway:ai-aws-guardrails-plugin
  name: AI AWS Guardrails Plugin
  description: Integrates Amazon Bedrock Guardrails for safety enforcement on traffic passing through Kong AI Gateway.
  humanURL: https://developer.konghq.com/plugins/ai-aws-guardrails/
  tags:
  - Plugin
  - Content Safety
  - AWS
  properties:
  - type: Documentation
    url: https://developer.konghq.com/plugins/ai-aws-guardrails/
- aid: kong-ai-gateway:ai-gcp-model-armor-plugin
  name: AI GCP Model Armor Plugin
  description: Integrates Google Cloud Model Armor for safety inspection on prompts and responses.
  humanURL: https://developer.konghq.com/plugins/ai-gcp-model-armor/
  tags:
  - Plugin
  - Content Safety
  - GCP
  properties:
  - type: Documentation
    url: https://developer.konghq.com/plugins/ai-gcp-model-armor/
- aid: kong-ai-gateway:ai-lakera-guard-plugin
  name: AI Lakera Guard Plugin
  description: Integrates Lakera Guard for prompt-injection and jailbreak detection.
  humanURL: https://developer.konghq.com/plugins/ai-lakera-guard/
  tags:
  - Plugin
  - Prompt Injection
  - Content Safety
  properties:
  - type: Documentation
    url: https://developer.konghq.com/plugins/ai-lakera-guard/
- aid: kong-ai-gateway:ai-semantic-response-guard-plugin
  name: AI Semantic Response Guard Plugin
  description: Inspects model responses by meaning and blocks responses that violate configured semantic policies.
  humanURL: https://developer.konghq.com/plugins/ai-semantic-response-guard/
  tags:
  - Plugin
  - Response Filtering
  - Semantic
  properties:
  - type: Documentation
    url: https://developer.konghq.com/plugins/ai-semantic-response-guard/
- aid: kong-ai-gateway:ai-custom-guardrail-plugin
  name: AI Custom Guardrail Plugin
  description: Lets operators define custom guardrail logic for prompts and responses without writing a full Kong plugin.
  humanURL: https://developer.konghq.com/plugins/ai-custom-guardrail/
  tags:
  - Plugin
  - Guardrails
  - Customization
  properties:
  - type: Documentation
    url: https://developer.konghq.com/plugins/ai-custom-guardrail/
- aid: kong-ai-gateway:ai-request-response-transformer-plugin
  name: AI Request/Response Transformer Plugin
  description: Uses LLMs at the gateway to transform request and response payloads (for example, reshaping JSON or translating
    fields) on the data path.
  humanURL: https://developer.konghq.com/plugins/ai-request-transformer/
  tags:
  - Plugin
  - Transformation
  - LLM
  properties:
  - type: Documentation
    url: https://developer.konghq.com/plugins/ai-request-transformer/
- aid: kong-ai-gateway:kong-agent-gateway
  name: Kong Agent Gateway
  description: Kong Agent Gateway is a capability of Kong AI Gateway (GA April 2026 with AI Gateway 3.14) that governs agent-to-agent
    (A2A) communication. It enforces agent identity verification, real-time policy and prompt-injection inspection, per-agent
    cost allocation, and unified observability across LLM calls, MCP tool invocations, and A2A messages.
  humanURL: https://konghq.com/blog/product-releases/kong-agent-gateway
  tags:
  - Agent Gateway
  - A2A
  - AI Gateway
  properties:
  - type: Documentation
    url: https://developer.konghq.com/ai-gateway/
  - type: Blog
    url: https://konghq.com/blog/product-releases/kong-agent-gateway
- aid: kong-ai-gateway:kong-mcp-registry
  name: Kong MCP Registry
  description: Kong MCP Registry (launched February 2026) is an enterprise directory inside Kong Konnect for registering,
    discovering, and governing MCP servers and AI-native tools. It provides dynamic discovery for AI agents, governance gates
    on which MCP resources are approved for use, and centralized observability of tool usage, health, and failures.
  humanURL: https://konghq.com/products/mcp-registry
  tags:
  - MCP
  - Registry
  - Konnect
  - Agentic AI
  properties:
  - type: Documentation
    url: https://developer.konghq.com/
  - type: Blog
    url: https://konghq.com/blog/product-releases/kong-mcp-registry-tech-preview
common:
- type: Portal
  url: https://konghq.com/products/kong-ai-gateway
- type: Documentation
  url: https://developer.konghq.com/ai-gateway/
- type: GettingStarted
  url: https://developer.konghq.com/ai-gateway/get-started/
- type: APIReference
  url: https://developer.konghq.com/ai-gateway/ai-providers/
- type: ChangeLog
  url: https://developer.konghq.com/gateway/changelog/
- type: Blog
  url: https://konghq.com/blog
- type: GitHubOrganization
  url: https://github.com/Kong
- type: GitHubRepository
  url: https://github.com/Kong/kong
- type: SDK
  url: https://github.com/Kong/sdk-konnect-go
  name: Kong Konnect Go SDK
- type: CLI
  url: https://github.com/Kong/kongctl
  name: Kong Developer CLI
- type: MCPServer
  url: https://github.com/Kong/mcp-konnect
  name: Kong Konnect MCP Server
- type: Pricing
  url: https://konghq.com/pricing
- type: LinkedIn
  url: https://www.linkedin.com/company/konghq
- type: Support
  url: https://discuss.konghq.com/
- type: Features
  data:
  - name: Universal LLM API
    description: One normalized request shape across 16+ providers (OpenAI, Anthropic, Azure AI, Bedrock, Gemini, Vertex AI,
      Cohere, Hugging Face, Llama, Mistral, xAI, DashScope, Cerebras, Ollama, Databricks, DeepSeek, vLLM).
  - name: Prompt Firewalls and PII Sanitization
    description: AI Prompt Guard, AI Semantic Prompt Guard, and AI PII Sanitizer block disallowed content and redact sensitive
      data before it hits the model.
  - name: Semantic Caching
    description: AI Semantic Cache serves semantically similar prompts from cache to cut latency and provider spend.
  - name: Token-Aware Rate Limiting
    description: AI Rate Limiting Advanced enforces per-consumer and per-model token budgets, not just request counts.
  - name: Automated RAG Injection
    description: AI RAG Injector pulls retrieved context into prompts at the gateway so application code stays simple.
  - name: Multi-Provider Load Balancing
    description: AI Proxy Advanced distributes traffic across providers with weighted distribution and fallback.
  - name: Content Safety Integrations
    description: Built-in integrations with Azure Content Safety, AWS Bedrock Guardrails, GCP Model Armor, and Lakera Guard.
  - name: MCP Traffic Governance
    description: Kong MCP Registry registers and governs MCP servers and tools that AI agents discover and call through the
      gateway.
  - name: Agent-to-Agent Governance
    description: Kong Agent Gateway (GA April 2026 with AI Gateway 3.14) governs A2A traffic with identity, policy, and observability.
  - name: Token-Level Observability
    description: Per-request, per-agent, and per-model token and cost telemetry surfaced into Konnect dashboards.
- type: UseCases
  data:
  - name: Provider-Agnostic LLM Access
    description: Give applications a stable LLM endpoint while swapping providers and models behind the gateway.
  - name: AI Cost Control
    description: Apply token budgets, semantic caching, and prompt compression to keep LLM spend bounded.
  - name: AI Safety and Compliance
    description: Enforce prompt firewalls, PII redaction, jailbreak detection, and content safety on every prompt and response.
  - name: Agentic Tool Governance
    description: Govern which MCP tools agents can discover and call, and inspect agent-to-agent traffic in production.
  - name: RAG at the Edge
    description: Inject retrieval context into prompts at the gateway without changing client code.
- type: Integrations
  data:
  - name: OpenAI, Anthropic, Azure AI, Bedrock, Gemini, Vertex AI
    description: Native provider integrations exposed through the AI Proxy plugin's normalized API.
  - name: Cohere, Hugging Face, Llama, Mistral, xAI, DashScope, Cerebras, Ollama, Databricks, DeepSeek, vLLM
    description: Additional native provider targets for AI Proxy and AI Proxy Advanced.
  - name: Lakera Guard
    description: Prompt-injection and jailbreak detection as a guardrail plugin.
  - name: Azure AI Content Safety / AWS Bedrock Guardrails / GCP Model Armor
    description: Cloud-native content safety integrations available as Kong AI plugins.
  - name: Prometheus, Grafana, OpenTelemetry
    description: Standard Kong Gateway observability stack carries LLM, MCP, and A2A telemetry.
- type: LLMsTxt
  url: https://developer.konghq.com/llms.txt
maintainers:
- FN: Kin Lane
  url: http://apievangelist.com
  email: kin@apievangelist.com