Agent-to-agent communication is the next frontier of AI infrastructure. As teams decompose monolithic AI workflows into specialized agents — a research agent, a booking agent, a summarization agent — the calls between those agents become as important to govern as the calls from end users. With…
[](/blog/product-releases/kong-ai-gateway-3-14)
Kong Agent Gateway is a new capability within Kong AI Gateway that extends our platform to more robustly cover agent-to-agent (A2A) communication. With this release, Kong AI Gateway now natively handles all three patterns of AI traffic in a product that is already being used by some of the largest…
[](/blog/product-releases/kong-agent-gateway)
The Demo-to-Production Gap Production reality multiplies complexity exponentially. In production, you need: Multi-tool orchestration with different authentication methods Robust retry logic with exponential backoff and circuit breakers Performance under load with multiple concurrent AI agents…
[](/blog/enterprise/build-vs-buy-mcp-server-infrastructure)
Traffic patterns shape architectural boundaries and understanding them clearly reveals why different tools are necessary. North-south traffic flows between external clients and your services, crossing the network perimeter in the process. Common examples of north-south traffic include: API gateways…
[](/blog/enterprise/the-difference-between-api-gateways-and-service-mesh)
The CLI case is real Let's start with the strongest version of the CLI argument. For well-known tools baked into model training data (e.g., git, grep, curl, jq, docker, kubectl, gh, etc.), CLIs are highly effective for agent workflows. The model already knows the flags and the output formats. And…
[](/blog/enterprise/cli-vs-mcp-enterprise-ai-governance)
In January 2024, consumers in the United Kingdom made a record-breaking 14.5 million open banking payments. This milestone shows how dramatically the financial services industry has changed. It's the result of years of regulatory work that kicked off back in 2017, when the UK's Competition and…
[](/blog/learning-center/guide-on-open-banking)
Showback and chargeback are not the same thing. Most organizations conflate these two concepts, and that conflation delays action. Understanding the LLM showback vs chargeback difference is the first step toward effective AI cost governance. Both are valuable, and each has its place given certain…
[](/blog/enterprise/llm-cost-management-ai-showback-and-chargeback)
The Stakes Keep Rising The security implications are severe. OWASP's 2025 Top 10 for LLM Applications ranks prompt injection as the number one critical vulnerability. Attackers manipulate LLM inputs to override instructions, extract sensitive data, or trigger unintended behaviors. Traditional…
[](/blog/engineering/5-best-practices-securing-microservices-scale)
The Anatomy of Architectural Complexity Modern architectures now juggle three distinct traffic patterns. Each brings unique demands. Traditional approaches treat them separately. This separation creates unnecessary complexity. North-South API Traffic External APIs are your digital front door.…
[](/blog/enterprise/microservices-to-ai-traffic-kong-as-the-unified-control-plane)
Traditional APIs are, in a word, predictable. You know what you're getting: AI APIs, especially anything that runs on LLMs under the hood, are a completely different story. These things are highly unpredictable by their very nature. You're dealing with token-based costs that swing wildly depending…
[](/blog/engineering/monetize-ai-apis)
Free collaboration with Postman — a myth On March 1st, 2026, Postman discontinued free collaboration for small teams. Now , Git or Cloud-native collaboration requires a Team plan starting at $19 per person per month. That means even a 3-person team is now paying $57 per month just to keep their API…
[](/blog/enterprise/insomnia-vs-postman-evaluating-api-testing-tools)
The widespread adoption of Kafka and event streaming platforms is evident across several enterprises, where they serve as the backbone of critical operations, ranging from financial transactions to AI inference pipelines. However, in the domains of security and identity, most teams continue to face…
[](/blog/product-releases/kong-event-gateway-1-1)
Kong does not rely on LiteLLM — whether PyPI-distributed or otherwise — for any components in our runtime stack. As you may know, a supply chain vulnerability affecting LiteLLM versions 1.82.7 and 1.82.8, a popular open-source AI proxy library, was publicized yesterday. The malicious package,…
[](/blog/news/kong-not-affected-by-the-pypi-distributed-litellm-supply-chain-attack)
Running Kong in front of your Solace Broker adds real benefits: Imagine an order service that accepts REST requests and publishes them as events. 1. API Consumer calls Kong with a standard POST request: 2. Kong Gateway applies authentication, validates the payload, adds a correlation ID, rate…
[](/blog/engineering/smarter-event-driven-apis-kong-solace)
For product teams, the long days of lengthy Product Requirements Documents (PRDs) and rigid documentation workflows are giving way to something faster and more dynamic. They’re now using AI to generate prototypes, validate concepts with virtual customers, and move from idea to working software in a…
[](/blog/news/gartner-recognizes-kong-progressive-ai-adopter)Secure your AI infrastructure with prompt guards, PII sanitization, and centralized governance. Control LLM costs with token-based rate limiting and semantic routing across providers.