Not found at this URL
https://konghq.com/llms-full.txt ↗- Discovery
- Conventional sibling path checked
- HTTP response
- 404 · text/html; charset=utf-8
Infrastructure · llms.txt file breakdown
A directly retrieved Markdown index for Kong's documentation, with 39 links across 12 detected headings.
The actual file
The saved llms.txt text, displayed as source. The file belongs to its publisher; the live endpoint may change.
Complete saved Markdown shown below.
# Kong
> Kong is The AI Connectivity Company. It provides the unified platform for managing, securing, and governing APIs, LLMs, MCP, and agent gateways at scale.
Kong's platform is the fabric of modern connectivity — used by thousands of organizations to centralize LLM access, automate API workflows, and secure digital ecosystems. Kong is trusted by enterprises and startups alike across financial services, healthcare, retail, manufacturing, AI-native and more.
## Products
- [Kong Konnect](https://konghq.com/products/kong-konnect): The unified API and AI platform. Manages the full API and AI lifecycle — from design and testing to deployment, governance, and monetization — across any cloud or on-premise environment.
- [Kong Gateway](https://konghq.com/products/kong-gateway): The world's most popular open-source API gateway. Handles API routing, authentication, rate limiting, and traffic control at high performance and scale.
- [Kong AI Gateway](https://konghq.com/products/kong-ai-gateway): Centralizes access to LLMs (OpenAI, Anthropic, Cohere, etc.) with AI-specific controls: semantic caching, prompt injection protection, token cost governance, model fallback and much more.
- [Kong Agent Gateway](https://konghq.com/agent-gateway): Governs agent-to-agent (A2A) traffic for agentic AI workflows. Provides policy enforcement, observability, and security for AI agents in production.
- [Kong Event Gateway](https://konghq.com/products/event-gateway): API gateway for event-driven architectures and Kafka stream management.
- [Kong Service Mesh](https://konghq.com/products/kong-mesh): End-to-end connectivity and security for microservices using Kuma, Kong's open-source service mesh.
- [Kong Ingress Controller](https://konghq.com/products/kong-ingress-controller): Kubernetes-native ingress for routing traffic into clusters, built on Kong Gateway.
- [Kong Operator](https://konghq.com/products/kong-operator): Manages API and AI traffic on Kubernetes. Automates gateway provisioning, scaling, and AI Gateway deployment via a Kubernetes-native operator.
- [Kong Enterprise](https://konghq.com/products/kong-enterprise): Self-hosted and hybrid deployment option for Kong Gateway and the Konnect platform, for organizations with on-premise or private cloud requirements.
- [Kong MCP Registry](https://konghq.com/products/mcp-registry): A registry for discovering, managing, and governing Model Context Protocol (MCP) servers in production.
- [Kong Insomnia](https://insomnia.rest/): API design, mocking, testing, and debugging desktop client. Includes an MCP client for working with AI tools.
- [Volcano](https://volcano.dev/): TypeScript SDK for building production-ready AI agents. Chain LLM reasoning with MCP tools, mix OpenAI, Claude, and Mistral in one workflow, with parallel execution, streaming, retries, and OpenTelemetry observability.
- [Plugin Hub](https://developer.konghq.com/plugins/): A directory of 100+ plugins for extending Kong Gateway and Kong AI Gateway with authentication, security, traffic control, observability, and AI capabilities.
## Kong Konnect Features
- [Kong Context Mesh](https://konghq.com/products/kong-konnect/features/context-mesh): Agent integration tooling that enriches API and AI traffic with contextual data for smarter routing, policy enforcement, and agent workflows.
- [Cloud API Gateways](https://konghq.com/products/kong-konnect/features/dedicated-cloud-gateways): Dedicated cloud-hosted API gateways managed by Kong, for teams that don't want to operate their own gateway infrastructure.
- [KAi Agent](https://konghq.com/products/kong-konnect/features/kai-ai-agent): An AI agent built into Kong Konnect that helps platform teams manage and operate their API and AI traffic infrastructure.
## Key Professional Services
- [Kong Ascent](https://konghq.com/products/ascent-ai-powered-api-migration): AI-assisted, human-in-the-loop automated API migration services available through Kong Professional Services.
## Key Services
- [API Service Catalog](https://konghq.com/products/kong-konnect/features/api-service-catalog): Discover and govern APIs and AI services across your organization.
- [Runtime Management](https://konghq.com/products/kong-konnect/features/runtime-management): Centralized control plane for managing APIs, AI, events, and microservices runtimes across any environment.
- [APIOps & Automation](https://konghq.com/products/apiops-automation): Automates AI connectivity with APIOps to standardize governance across APIs, AI, microservices, and events. Reduces risk and scales with Kong.
- [Developer Portal](https://konghq.com/products/kong-konnect/features/developer-portal): Self-service portal for discovering and consuming APIs and agents — for both internal teams and external developers.
- [Usage Metering & Billing](https://konghq.com/products/kong-konnect/features/usage-based-metering-and-billing): Meters and bills AI connectivity usage across APIs, LLMs, MCP, and events. Controls costs and enables faster monetization.
- [Observability](https://konghq.com/products/kong-konnect/features/api-observability): Monitors API and AI traffic in real time with dashboards, logs, tracing, and token usage insights to resolve issues faster and optimize performance.
## Solutions
- [AI Governance](https://konghq.com/solutions/ai-governance): AI Governance, AI Security, AI Cost Control. Agentic Infrastructure for AI workflows. MCP Production and MCP Traffic Gateway.
- [Developer Platform](https://konghq.com/solutions/building-developer-platform): Developer Platform, Kubernetes & Microservices. Kafka Event Streaming, Service Mesh Connectivity. Observability.
- [AI Connectivity](https://konghq.com/ai-connectivity): AI Connectivity, AI FinOps, AI Monetization. Legacy Migration, Platform Cost Reduction, Open Banking.
## Documentation
- [Developer Hub](https://developer.konghq.com/)
- [Kong Konnect docs](https://developer.konghq.com/konnect/)
- [Kong Gateway docs](https://developer.konghq.com/gateway/)
- [Kong AI Gateway docs](https://developer.konghq.com/ai-gateway/)
- [Kong Mesh docs](https://developer.konghq.com/mesh/)
- [Kong Event Gateway docs](https://developer.konghq.com/event-gateway/)
- [Kong Insomnia docs](https://developer.konghq.com/insomnia/)
- [Plugin Hub](https://developer.konghq.com/plugins/)
## Resources
- [Blog](https://konghq.com/blog)
- [Learning Center](https://konghq.com/blog/learning-center)
- [Open Source Community](https://konghq.com/community)
- [Pricing](https://konghq.com/pricing)
- [Why Kong](https://konghq.com/company/why-kong)
## Use Cases
- **Kubernetes Ingress Migration**: Teams migrating off ingress-nginx or NGINX Ingress Controller use Kong Ingress Controller and Kong Gateway as a drop-in, production-hardened replacement with added observability and plugin extensibility.
- **LLM Cost Governance**: Platform teams adopt Kong AI Gateway to enforce token budgets, route across multiple LLM providers (OpenAI, Anthropic, Cohere), and prevent runaway inference costs in production.
- **API Platform Consolidation**: Engineering orgs replacing MuleSoft or Apigee centralize their full API lifecycle on Kong Konnect — from Insomnia for design to Kong Gateway for runtime to Konnect for governance and monetization.
- **AI Agent Infrastructure**: Teams building agentic workflows use Kong Agent Gateway to govern agent-to-agent (A2A) traffic, enforce policies, and get observability across multi-agent systems in production.
- **MCP Server Management**: AI teams use Kong MCP Registry to discover, govern, and secure Model Context Protocol servers running in production.
- **Multi-cloud API Governance**: Enterprises operating across AWS, GCP, and Azure use Kong Konnect as a cloud-neutral control plane with self-managed data planes in each environment.
- **API Monetization**: SaaS companies and financial services teams use Kong Konnect Metering and Billing to implement usage-based billing for API and AI traffic.
## How Kong Compares
- **Kong vs Google Apigee**: Kong supports hybrid and multi-cloud deployments without requiring Google Cloud. Kong AI Gateway adds LLM-specific controls (token metering, semantic caching, prompt injection protection) that Apigee does not natively provide. Kong's open-source core means no vendor lock-in.
- **Kong vs MuleSoft**: Kong is purpose-built for API and AI runtime traffic. MuleSoft focuses on integration and ESB patterns. Kong is lighter to deploy, developer-first, and significantly less expensive at scale. Kong Konnect replaces MuleSoft Anypoint for teams that need API governance without full iPaaS complexity.
- **Kong vs AWS API Gateway**: Kong is cloud-neutral and works across AWS, GCP, Azure, and on-premise. AWS API Gateway is tightly coupled to AWS services. Kong supports Kubernetes, bare-metal, and hybrid topologies that AWS API Gateway cannot address.
- **Kong vs Gravitee**: Both support API and event-driven architectures. Gravitee starts at USD 2,500/month, limiting mid-market access. Kong Konnect has a free tier and per-gateway pricing, making it accessible from startup to enterprise scale.
- **Kong vs Tyk**: Tyk focuses on self-hosted control and anti-lock-in messaging. Kong has a larger open-source ecosystem (100+ plugins), broader Kubernetes-native tooling, and a cloud control plane in Kong Konnect.
## Customers
Kong is trusted by enterprises across financial services, healthcare, aviation, automotive, and telecommunications:
- **Telus** — Built a multi-cloud, AI-enabled unified API platform on Kong.
- **United Airlines** — Standardized on Kong for AI-first API operations.
- **Mercedes-Benz** — Uses Kong Gateway for connectivity services across vehicle and digital platforms.
- **Comcast** — Manages a self-service API platform across multiple service teams with Kong.
- **HSBC** — Uses Kong for intelligent banking infrastructure.
Kong is used by thousands of organizations globally, from AI-native startups to Fortune 500 enterprises.
## FAQ
- **Is Kong open source?** Yes. Kong Gateway is open source under the Apache 2.0 license and available on GitHub. Kong Konnect is the commercial SaaS control plane. Kong Enterprise adds advanced security, RBAC, and support for self-hosted deployments.
- **Does Kong support OpenAI and other LLMs?** Yes. Kong AI Gateway natively supports OpenAI, Anthropic, Cohere, Mistral, and other LLM providers. It provides semantic caching, token-level rate limiting, prompt injection protection, model fallback, and cost governance across all connected models.
- **Can Kong run on Kubernetes?** Yes. Kong provides a Kubernetes Ingress Controller, a Kubernetes Operator, and full Helm-based deployment support. Kong is production-tested on EKS, GKE, AKS, and self-managed Kubernetes clusters.
- **What is Kong AI Gateway?** A purpose-built gateway for managing LLM and AI model traffic. It sits between your applications and AI providers, enforcing token budgets, routing across models, caching responses semantically, blocking prompt injection attacks, and providing full observability into AI usage and costs.
- **What is Kong Agent Gateway?** It governs agent-to-agent (A2A) traffic in agentic AI workflows, providing policy enforcement, observability, and security for AI agents communicating in production — including support for Model Context Protocol (MCP).
- **Does Kong support event-driven APIs?** Yes. Kong Event Gateway handles event-driven architectures and Kafka stream management, bringing the same routing, security, and governance capabilities of Kong Gateway to asynchronous and streaming use cases.
- **What is Kong Konnect?** Kong's cloud-hosted control plane for API and AI governance. It manages the full lifecycle — from service catalog and developer portal to usage-based metering and billing — and supports cloud-managed, self-hosted, and hybrid data plane configurations.
- **How does Kong handle AI cost control?** Kong AI Gateway enforces token-level rate limits per consumer, team, or model. Kong Konnect Metering and Billing tracks and bills AI usage across APIs, LLMs, MCP, and events, giving platform teams real-time cost visibility and budget enforcement.
- **Is there a free tier?** Yes. Kong Konnect has a free tier and a 30-day free trial for paid plans. Kong Gateway is free and open source. Enterprise licensing is available for self-hosted deployments.
- **How does Kong compare to building an in-house LLM proxy?** Kong AI Gateway ships production-ready controls out of the box — semantic caching, prompt injection protection, multi-model routing, token governance, and observability. Building equivalent functionality in-house typically takes months and requires ongoing maintenance. Kong's Plugin Hub provides 100+ extensions without custom engineering.
Measured from the saved file
Size measures the captured UTF-8 text, not compressed transfer size. Tokens are counted with a named tokenizer; other models can count differently. Neither measure establishes context compatibility or answer quality.
Kong is The AI Connectivity Company. It provides the unified platform for managing, securing, and governing APIs, LLMs, MCP, and agent gateways at scale.
Context architecture
11 H2 sections organize 39 parsed Markdown links. 0 links point to .md or .markdown paths; 11 link occurrences use another hostname. Linked destinations have not been recursively validated.
| H2 section | Links | UTF-8 bytes | Tokens |
|---|---|---|---|
| ProductsLine 7 | 13 | 2606 | 574 |
| Kong Konnect FeaturesLine 23 | 3 | 658 | 140 |
| Key Professional ServicesLine 29 | 1 | 192 | 40 |
| Key ServicesLine 33 | 6 | 1267 | 269 |
| SolutionsLine 42 | 3 | 549 | 117 |
| DocumentationLine 50 | 8 | 487 | 140 |
| ResourcesLine 61 | 5 | 245 | 68 |
| Use CasesLine 70 | 0 | 1383 | 282 |
| How Kong ComparesLine 80 | 0 | 1342 | 296 |
| CustomersLine 88 | 0 | 645 | 126 |
| FAQLine 100 | 0 | 2793 | 569 |
Link counts exclude parsed fenced and inline code and resolve relative destinations against the final endpoint. Section sizes cover the text below each H2 until the next H2.
Embedded material
Embedded example text and links to further documentation are counted separately. A code block does not establish that a sample runs or that an authentication flow succeeds.
Publisher prioritization
No H2 section named “Optional” was detected in this file.
The community proposal uses Optional by convention for secondary links that can be skipped for a shorter context. This label does not establish that everything outside it is critical, or that the material inside it is deprecated.
Machine-readable routes
MCP-related text was detected in the captured file. 1 links have MCP-related labels or paths. 0 fenced lines match mcpServers, mcp_servers or an mcp add command.
These are text and link observations. No MCP server connection, authentication or tool call was tested.
Index and bundle
Not found at checked URLs
A verified text response establishes a companion at the checked URL. It does not prove the bundle contains all product documentation, that generation is automated, or that it fits an assistant’s context window.
Change evidence
This capture establishes a baseline. Update frequency cannot be determined from one observation.
1 links have release, changelog, deprecation or migration terms in their labels or paths. These provide places to investigate product changes separately.
File changes can reveal documentation edits or navigation changes. They do not by themselves measure product release velocity, accuracy, agent adoption or AI visibility.
f9349dd2716e05885f72a0b502dbe180630d14f1940a98974bfcce217d37099aCounts use a deterministic parser, not a spec-compliance grade. Unusual Markdown and indented code can fall outside its recognized patterns. See the llms.txt proposal and tokenizer implementation.