Native capabilities and the reference implementation

SOLO ENTERPRISE FOR

Agentic Infrastructure

agentgateway agentregistry kagent

Cloud Connectivity

Istio kgateway

Get started with Solo

Get Started

Get started with Solo

Get Started

Get started with Solo

Get Started

AI costs are easy to underestimate until agents, copilots, evaluations, and automated pipelines begin calling multiple models at once. Unlike a conventional API request, an LLM request does not map neatly to a predictable unit of work. Cost depends on the selected model, prompt size, generated output, retries, tool calls, and provider-specific token accounting.

Provider dashboards remain useful, but they are usually retrospective. Platform teams need an enforcement point that can stop runaway consumption before the next provider request is made. The gateway is a natural place to do that: every request already passes through it, so the gateway can attach trusted identity, reserve budget capacity, record actual token usage, and reject calls when an allocation is exhausted.

This post shows how to build a dollar-denominated AI cost-governance layer on top of agentgateway's native External Processing support. The reference implementation adds hierarchical budgets, pre-request reservations, actual-cost settlement, forecasting, approvals, rate limits, and a management UI.

The design follows four steps:

Validate the caller and overwrite trusted identity headers.
Reserve estimated budget before the LLM request leaves the gateway.
Settle the reservation using provider-reported token usage.
Alert, forecast, and optimize using real spend data.

The complete example used in this post, including deployment configuration and seed data, is available at github.com/day0ops/quota-management.

Native capabilities and the reference implementation

Before getting into the implementation, it is important to distinguish between the capabilities provided by agentgateway and the additional controls implemented by the example project.

Agentgateway provides the in-path extension point. Its External Processing support allows a gRPC service to inspect request and response traffic, mutate headers or bodies, and return an immediate response before traffic reaches an upstream provider. Solo Enterprise for agentgateway also provides native request and token rate-limiting patterns, along with LLM cost-tracking metrics.

The reference implementation in this post builds a richer dollar-denominated control plane on top of those capabilities.

For simpler use cases, native token budgets may be enough. The reference implementation becomes useful when teams need financial ceilings, delegated ownership, forecasting, and a clear approval trail.

Useful background reading:

Why AI spend needs gateway control

AI spend is not just another API bill. It is variable, fast-moving, and tied to behavior that can change from one request to the next.

Runaway agent loops

An agent that retries after an error or gets stuck in a tool-call cycle can send hundreds of requests in seconds. Each request may include an increasingly large conversation history. Without a hard stop, one bad loop can consume a team's monthly allocation before anyone sees an alert.

Context growth

A task that starts with a small prompt can grow into a large context window over a long session. Input and output tokens are priced differently, and output-heavy workflows can dominate the bill.

Multi-model sprawl

Teams often use OpenAI, Anthropic, Google, local models, hosted embeddings, and provider-specific fallbacks at the same time. Without per-team and per-model attribution, platform teams cannot tell whether a budget increase is justified or whether a cheaper model would do the job.

Blunt provider limits

Provider-side spend limits are useful as a final backstop, but they are usually account-level. They do not map cleanly to orgs, teams, users, projects, environments, or internal approval workflows.

Good budget control turns cost into an in-path policy decision. Requests carry trusted identity. Budgets are scoped to the organization. Usage is logged by model. Alerts and forecasts appear while there is still time to act.

A concrete customer scenario

Consider an organization called Acme with a monthly AI budget of $10,000.

A developer-facing copilot and a production support agent can share the same gateway while receiving different policies. If an evaluation job enters a runaway loop, the gateway rejects the next provider request when the $500 allocation is exhausted. The support agent remains unaffected because its traffic is charged to a different budget.

That is the key benefit of gateway enforcement: the control point is shared, but the policy is not blunt.

Architecture

The reference implementation has four main components:

‍

High-level architecture showing agentgateway, the budget ExtProc service, the rate-limit ExtProc service, PostgreSQL, the management UI, and upstream LLM providers

The budget service participates synchronously in the request path. This makes it suitable for enforcement rather than delayed reporting.

How enforcement works

Every LLM request passes through two budget stages:

Pre-request reservation before the provider call begins.
Post-response settlement after token usage is available.

Sequence diagram showing a pre-request budget reservation and a post-response cost settlement

Why reservation and settlement are both necessary

An LLM request does not have a fully known cost before it begins. Output tokens depend on what the model generates. Prompt tokens may also be unavailable at the earliest enforcement point unless they are estimated or derived before forwarding.

A reservation prevents the gateway from allowing unlimited requests while waiting for actual cost data. Settlement replaces the estimate with the provider-reported usage after the response is complete.

Doing only one of these creates problems:

Settlement without reservation allows concurrent requests to exceed the remaining budget before accounting catches up.
Reservation without settlement blocks requests based on imprecise estimates and never records the true spend.

The enforcement decision belongs in `RequestHeaders`

The hard blocking decision is made during the RequestHeaders phase, before request-body streaming begins. Once body chunks have started moving upstream, returning a reliable immediate rejection becomes more complicated.

The gateway must therefore normalize the model name and attach trusted identity before invoking the budget ExtProc service.

A practical reservation strategy looks like this:

ExtProc phases

`RequestHeaders`

The service reads gateway-stamped request context such as:

x-gw-org-id
x-gw-team-id
x-gw-user-id
x-gw-llm-model
x-gw-environment
x-gw-request-id
Optional metadata used for budget matching

It evaluates budget match expressions, estimates cost from the model-pricing table, and atomically reserves budget capacity.

There are two possible outcomes:

ImmediateResponse with a configurable HTTP status when the budget is exhausted.
HeadersResponse with CONTINUE when budget is available.

`RequestBody`

The ExtProc service receives request-body chunks. It may reconstruct the body for validation, telemetry, or debugging, but the body should not be the primary hard-enforcement point.

Avoid persisting prompts unless there is a clear requirement. Usage metadata is generally enough for cost governance.

`ResponseHeaders`

Provider response headers pass through. The service may attach correlation metadata if needed.

`ResponseBody`

At end of stream, the service parses token usage, calculates actual cost, records usage, releases the reservation, and attaches response headers such as:

x-budget-cost-usd
x-budget-remaining-usd
x-budget-id

Copy

Copied!

The parser handles common provider formats:

Calculating cost

Once token counts are extracted, cost is calculated using per-model pricing stored in PostgreSQL:

cost = (input_tokens / 1_000_000) × input_cost_per_million
     + (output_tokens / 1_000_000) × output_cost_per_million

Copy

Copied!

For illustration, assume a small model costs $0.15 per million input tokens and $0.60 per million output tokens. A request using 500 input tokens and 800 output tokens costs:

(500 / 1_000_000) × 0.15 + (800 / 1_000_000) × 0.60
= $0.000075 + $0.000480
= $0.000555

Copy

Copied!

The same formula applies during reservation, but the token counts are estimates. The implementation should use conservative defaults, requested output-token caps, and a configurable multiplier to reduce the risk of under-reserving.

Treat pricing as versioned configuration

Model prices change over time. Do not hard-code them into application logic.

Each usage record should store:

Requested model ID
Matched catalog model ID
Provider
Input and output token counts
Input and output unit prices applied at settlement time
Total cost
Pricing-table version or effective date
Match strategy, such as exact, normalized, or fallback

This makes historical usage auditable even after prices change.

Identity and the trust boundary

Every request carries a JWT issued by an identity provider. Agentgateway validates the token before routing and overwrites trusted headers for ExtProc.

The important word is overwrites. The gateway must never trust identity headers supplied by the client.

An illustrative transformation looks like this:

yamltransformation:
  request:
    set:
      - name: x-gw-org-id
        value: "jwt['org_id']"
      - name: x-gw-team-id
        value: "jwt['team_id']"
      - name: x-gw-user-id
        value: "jwt['sub']"

Copy

Copied!

The exact policy resource depends on the agentgateway deployment mode and version. The principle remains the same:

Validate the JWT.
Remove or overwrite caller-provided identity headers.
Normalize the model name.
Attach a correlation ID.
Invoke ExtProc only after the request context is trusted.

Use a small, consistent set of fields:

Keep these fields boring, consistent, and visible. Then build dashboards that answer who spent money, on which model, in which environment, and why.

Budget model

Budgets are scoped to orgs, teams, and users.

Budget hierarchy showing organization, team, and user allocations

Each budget has a CEL match expression. A budget can match on org, team, user, model, request headers, gateway metadata, or any combination.

Isolation and fallback

Child budgets support two useful behaviors.

Isolated budgets are hard caps. If an isolated team budget is exhausted, the request is rejected even when the org budget still has capacity. Use this for sandbox environments, test pipelines, or projects that must not exceed their allocation.

Fallback budgets can draw from the parent when the child is exhausted. If the team budget is empty but the org budget has room, the request is allowed and the usage record marks parent_charged: true. If the parent is also exhausted, the request is rejected.

Diagram showing isolated and fallback budget behavior

Periods and retry behavior

Budgets can reset hourly, daily, weekly, monthly, or on a custom interval. A background worker resets expired budgets by clearing period counters and advancing current_period_start.

Rejected responses should include a machine-readable body and, where appropriate, a Retry-After header containing the number of seconds until reset.

Example rejection:

{
  "error": "budget_exhausted",
  "budget_id": "engineering-monthly",
  "scope": "team",
  "remaining_usd": "0.000000",
  "resets_at": "2026-07-01T00:00:00Z",
  "message": "The AI budget for this team has been exhausted. Stop the current task and surface this error to the user."
}

Copy

Copied!

The status code should be configurable because automated clients do not all handle retryable errors in the same way.

Approval workflow

Budget creation requires approval before enforcement. New budgets start in pending state. Org admins can approve or reject budgets in their org. Team admins can manage budgets within their permitted scope.

Budget approval workflow from creation to approval or rejection

Rejections should require a written reason. Resubmissions should remain auditable. The same workflow can apply to rate-limit allocations.

Here's an example of the budget approval workflow on the management UI:

Admin roles and permissions

The management API enforces role-based access at every write operation.

Only org admins should be able to expand budgets that sit under an organization ceiling. This prevents teams from quietly increasing their own allocation without an approval trail.

Model pricing and unknown models

The ExtProc service stores model prices in PostgreSQL. Each record contains input and output cost per million tokens. Optional fields cover cache-read and cache-write pricing for providers that support prompt caching.

Model pricing catalog showing model IDs, providers, and unit costs

Model lookup may normalize provider-specific version suffixes. For example, a dated model identifier may be matched to a stable pricing-catalog entry. Any fuzzy or normalized match must be auditable.

Use a safe unknown-model policy

An unknown model should not silently inherit an optimistic default price in production. That can allow an expensive new model to consume more budget than expected.

Make the behavior configurable:

Log the requested model ID, matched catalog key, match strategy, and applied rate for every request.

Production hardening

A synchronous enforcement service becomes part of the critical request path. The failure policy must be deliberate.

Agentgateway supports failClosed and failOpen behavior for ExtProc failures. For governed LLM routes, prefer failClosed: if the budget service cannot make a reliable decision, block the request rather than silently bypassing controls.

Security guidance

ExtProc can inspect request and response bodies. In an LLM gateway, those bodies may include prompts, generated content, tool inputs, and sensitive data.

Use these controls:

Encrypt traffic between agentgateway and the ExtProc service.
Do not store raw prompts or responses unless there is a clear requirement.
Store usage metadata separately from content.
Redact secrets from application and audit logs.
Restrict management UI access through OIDC and role-based authorization.
Store PostgreSQL credentials in Kubernetes secrets or an external secret manager.
Audit budget, role, and model-pricing changes.
Set retention periods for usage records and audit logs.

Operating the system

Once enforcement exists, day-to-day work shifts to visibility, forecasting, and optimization. Operators need to know which budgets are healthy, which teams are trending toward exhaustion, and where model choices can be improved.

Alerts and forecasting

The management UI calculates budget health continuously:

Four alert types fire automatically:

Forecast alerts fire before exhaustion. A budget with a $10 daily limit that has spent $2 in the first hour should trigger a warning immediately, long before the hard cap is reached.

Forecast screen showing projected spend and exhaustion estimates

Alerts screen showing threshold, exhaustion, and forecast notifications

A practical policy is:

Alert owners at 60% usage.
Escalate at 85% when burn rate still points to exhaustion.
Enforce at 100% and disable or degrade noncritical paths.

Budget detail and management UI

Each budget has a detail view showing current usage, limit, remaining capacity, reset period, hierarchy flags, and usage history.

Budget detail screen showing allocation, remaining spend, and request-level charges

The UI gives operators one place to manage:

Budget status and hierarchy flags.
Request-level usage history with concrete model IDs and token counts.
Forecasting based on real burn rate.
Threshold and exhaustion alerts.
Per-user and service-account assignments.
Approval queues.
Audit logs containing actor, action, timestamp, and metadata.
The model-pricing catalog.

The UI is intended for shared ownership. Platform teams enforce global controls, while team owners inspect their own usage and justify changes with data.

Observability

The budget service exposes Prometheus metrics on port 9090.

These metrics integrate with Prometheus and Grafana without requiring every application team to add custom instrumentation.

Complement budgets with rate limits and token caps

A spend budget limits cumulative financial exposure. It does not replace rate limiting.

A runaway agent can still generate a burst of concurrent requests before settlement catches up. Layered controls reduce that risk.

The reference implementation includes a separate quota-ratelimit-extproc service. Rate-limit allocations can be scoped per org, team, and model pattern.

Start new allocations in monitoring mode, observe real traffic, and switch to enforced after calibrating the limits.

You can also centrally cap provider request fields such as maximum output tokens or reasoning effort through gateway transformations. That reduces the worst-case cost of an individual request before it reaches the model provider.

Minimum adaptation path

The repository contains the full deployment manifests and seed data. The following sequence gives customers a practical mental model for adapting the design.

1. Deploy the data store

Deploy PostgreSQL and create tables for:

Budgets
Reservations
Usage records
Model prices
Alerts
Approval requests
Audit records

2. Deploy the budget ExtProc service

Configure the service with:

PostgreSQL connection details
Reservation TTL
Estimation defaults and multiplier
Unknown-model policy
Budget-exceeded status code
Pricing-cache TTL

3. Validate identity at the gateway

Configure JWT validation and overwrite the trusted context headers:

x-gw-org-id
x-gw-team-id
x-gw-user-id
x-gw-llm-model
x-gw-environment
x-gw-request-id

Copy

Copied!

4. Attach ExtProc with a fail-closed policy

Use the ExtProc policy supported by your agentgateway deployment mode. Verify that requests fail safely when the budget service is unavailable.

5. Seed model prices

Add the model IDs used by your organization. Include provider-specific variants, cache pricing where relevant, and an explicit unknown-model policy.

6. Create initial budgets

Start with a small number of easy-to-understand allocations:

One organization budget
One isolated sandbox budget
One production team budget with controlled fallback
One service-account budget for automated evaluations

7. Test allowed traffic

Send a request with a valid token and verify:

The upstream provider receives the request.
The usage record includes the org, team, user, model, and request ID.
The reservation is released after settlement.
The cost response headers are attached.

8. Test rejection behavior

Exhaust a small test budget and verify:

The provider is not called.
The gateway returns the expected machine-readable error.
The client behavior is acceptable.
Alerts and audit records appear.

9. Test failure modes

Deliberately stop the ExtProc service and PostgreSQL in a non-production environment. Confirm that the gateway follows the intended fail-closed policy and that orphaned reservations are reconciled.

Example: Claude Code through the gateway

The same enforcement path applies to developer tools and production services. Anything routed through agentgateway receives the same identity, budget checks, accounting, and telemetry.

Claude Code supports routing through an LLM gateway by setting a custom Anthropic-compatible endpoint and bearer token:

export ANTHROPIC_BASE_URL="https://<gateway-host>"
export ANTHROPIC_AUTH_TOKEN="<gateway-issued-token>"

Copy

Copied!

The gateway endpoint must expose the Anthropic Messages-compatible API expected by Claude Code. The bearer token reaches agentgateway as an Authorization: Bearer header and can be validated before the request is forwarded.

Claude Code request flow through agentgateway and the budget service

From the operations side, Claude Code usage appears beside other LLM traffic in the same UI. There is one budget model, one forecast, and one audit trail.

Validate retry behavior for your client version

Do not assume every client handles budget exhaustion in the same way. A 429 Too Many Requests response is useful when a retry after reset is appropriate. A permanently exhausted task may require a non-retryable response so the client stops and surfaces the error.

Make BUDGET_EXCEEDED_STATUS_CODE configurable and test the behavior against the Claude Code version deployed in your environment. Client retry behavior can evolve over time.

Optimization after enforcement

Once cost is visible and bounded, teams can optimize safely.

Route by task shape

Use smaller models for extraction, classification, formatting, and other structured outputs. Save large, long-context, or reasoning models for work that actually needs them.

Shadow cheaper models

Send a small percentage of traffic to a cheaper candidate model while logging quality and cost. Expand only when acceptance rate and latency hold up.

Cache repeat work

Cache stable prompts, normalized parameters, and deterministic tool outputs. Track cache hit rate next to cost per request.

Track model usage

Record usage by concrete model ID so operators can detect when a workflow begins using a more expensive model than expected. For broader grouping, use model naming conventions or patterns in the pricing catalog.

The operating sequence is straightforward:

Unify request identity and usage logging.
Apply budgets and rate limits.
Observe actual traffic.
Improve routing with evidence.

Summary

Gateway budget control gives teams enforcement instead of delayed reporting. Agentgateway provides the in-path External Processing integration point. The reference implementation builds on that foundation by reserving estimated cost before the provider call, settling actual usage after the response, and tying spend to trusted identity and model data.

The most important design principles are:

Validate identity and overwrite trusted context headers at the gateway.
Reserve budget before forwarding the provider request.
Settle with provider-reported token usage.
Version model pricing and reject unknown models safely.
Combine budgets with request limits, token limits, and per-request caps.
Fail closed when the cost-control service cannot make a reliable decision.
Test client retry behavior rather than assuming it.

That combination gives finance a clear ceiling, gives platform teams a shared control point, and gives application teams enough visibility to choose cheaper models, cache repeated work, and request budget increases with evidence.

Learn more about agentgateway

Get started with agentgateway
Check out the project on GitHub
Join the community on Discord
Learn more on the website

Featured content

\ \ Solo Academy Course: Introduction to Envoy Proxy \ \ Solo Academy | Hands-on workshop introducing the core concepts of Envoy Proxy. Learn how Envoy works under the hood from its architecture, filter chains, and request flow and beyond the abstractions of service meshes and API gateways.\ \ \ \ Read Lab](/content/resources/lab/solo-academy-course-introduction-to-envoy-proxy/index.html)

\ \ Gartner® Report: How to Adapt Your API Strategy to Succeed in the AI Era \ \ As organizations unlock the potential of generative AI, one thing is clear: a modern, scalable API strategy is essential. In this complimentary Gartner® report learn how software engineering leaders can evolve their API programs to accelerate innovation, reduce risk, and control costs in the AI era.\ \ \ \ Read Report](/content/resources/report/gartner-report-how-to-adapt-your-api-strategy-to-succeed-in-the-ai-era/index.html)

\ \ Choosing the Right AI Gateway For You \ \ Read our white paper and learn how to select the right AI Gateway to help you navigate the challenges of working with AI workloads. \ \ \ \ Read Whitepaper](/content/resources/white-paper/choosing-the-right-ai-gateway-for-you/index.html)

\ \ \ \ kagent <3 Agent Substrate: A 101 installation & Configuration Guide \ \ Learn how to run AI agents on Kubernetes with Agent Substrate and kagent. This step-by-step guide covers architecture, installation, configuration, and deploying agent workloads using the Substrate runtime.\ \ \ \ Read Blog](/content/blog/kagent-3-agent-substrate-a-101-installation-configuration-guide/index.html)

\ \ \ \ Solo Enterprise for Istio 1.30: Agentic Mesh, ztunnel-Native Egress, New UI, and Fine-Grained Workload Identity \ \ This release brings agentgateway into the ambient mesh as a waypoint, ingress, and egress proxy, adds egress controls directly in ztunnel, ships a revamped Solo UI with a new advanced Service Graph, and introduces workload identity that goes beyond the service account.\ \ \ \ Read Blog](/content/blog/solo-enterprise-for-istio-1-30-agentic-mesh-ztunnel-native-egress-new-ui-and-fine-grained-workload-identity/index.html)

\ \ \ \ Agentgateway Code Mode for OpenAPI to MCP \ \ A very common pattern for MCP servers in the enterprise is to wrap existing APIs. We’ve invested 20+ years in building useful, reusable, valuable APIs, should we really be reinventing everything just because a fancy new protocol showed up? Is that even practical? The answer is no. We need to be practical and recognize some fundamental challenges mapping APIs to MCP tools. With agentgateway, we have three ways to expose OpenAPI operations via MCP: direct exposure, custom exposure with optional API chaining, or an AI model-controlled “code mode”.\ \ \ \ Read Blog](/content/blog/agentgateway-code-mode-for-openapi-to-mcp/index.html)

\ \ \ \ From Service Mesh to Agentic Mesh \ \ Service mesh became boring...but it's more foundational than ever before.\ \ \ \ Read Blog](/content/blog/from-service-mesh-to-agentic-mesh/index.html)

\ \ \ \ Keeping Context and Tokens Low With Progressive Disclosure In Agentgateway \ \ Learn how to cut MCP token usage by 91% using agentgateway’s progressive disclosure. Reduce cost, control context bloat, and optimize agent workflows.\ \ \ \ Read Blog](/content/blog/keeping-context-and-tokens-low-with-progressive-disclosure-in-agentgateway/index.html)

\ \ \ \ MCP Progressive Disclosure: Save Tokens, Retrieve Schemas \ \ \ \ Read Blog](/content/blog/mcp-progressive-disclosure/index.html)

\ \ \ \ Five Minutes to Your First MCP Server Tool: A Quickstart with agentgateway \ \ New to agentic AI? This guide walks you through running agentgateway locally, connecting to MCP servers, and understanding core concepts like rate limiting and observability.\ \ \ \ Read Blog](/content/blog/five-minutes-to-your-first-mcp-server-tool-a-quickstart-with-agentgateway/index.html)

\ \ \ \ Agentic Quality Benchmarking With Agentevals \ \ \ \ Read Blog](/content/blog/agentic-quality-benchmarking-with-agent-evals/index.html)

\ \ \ \ The AppMesh Migration Playbook \ \ \ \ Read Blog](/content/blog/the-app-mesh-migration-playbook/index.html)

\ \ \ \ Solo Enterprise for Istio 1.29: ECS Now GA, Enhanced Debuggability, and Flexible Global Service Aliasing \ \ The latest Solo Enterprise for Istio release delivers General Availability of AWS ECS integration and powerful new global service aliasing capabilities.\ \ \ \ Read Blog](/content/blog/solo-enterprise-for-istio-1-29/index.html)

\ \ \ \ Your First AI Route: Connecting to OpenAI with AgentGateway \ \ \ \ Read Blog](/content/blog/your-first-ai-route-connecting-to-openai-with-agentgateway/index.html)

\ \ \ \ Getting started with Multi-LLM provider routing \ \ \ \ Read Blog](/content/blog/getting-started-with-multi-llm-provider-routing/index.html)

\ \ \ \ What Comes After Ingress NGINX? A Migration Guide to Gateway API \ \ Ingress NGINX is being retired. This guide walks through migrating Ingress configs to Kubernetes Gateway API using ingress2gateway and kgateway.\ \ \ \ Read Blog](/content/blog/what-comes-after-ingress-nginx-a-migration-guide-to-gateway-api/index.html)

\ \ \ \ Why Traditional Gateways Failed AI Workloads - and How Kgateway's Rust-powered Agentgateway Fixes It \ \ Most AI gateways patch legacy proxies. Kgateway starts from first principles, rethinking the gateway for the agentic era with a data plane built specifically for modern AI traffic.\ \ \ \ Read Blog](/content/blog/why-traditional-gateways-failed-ai-workloads-and-how-kgateways-rust-powered-agentgateway-fixes-it/index.html)

\ \ \ \ Context-aware Security for Agentic AI Gateways \ \ If your gateway can’t tell the difference between a tool call and a model invocation, it can’t enforce meaningful security. Agentic systems demand a new class of context-aware, AI-native gateways.\ \ \ \ Read Blog](/content/blog/context-aware-security-ai-gateways/index.html)

\ \ \ \ Kgateway: The Best Alternative for Ingress NGINX \ \ Learn how kgateway, a CNCF-hosted project built on Envoy, offers a trusted path forward for Ingress NGINX.\ \ \ \ Read Blog](/content/blog/kgateway-the-best-alternative-for-ingress-nginx/index.html)

\ \ \ \ The Linux Foundation’s new Agentic AI Foundation and Secure MCP Infrastructure \ \ \ \ Read Blog](/content/blog/aaif-announcement-agentgateway/index.html)

\ \ \ \ Security Holes in MCP Servers and How To Plug Them \ \ Learn how to close the major security gaps in Model Context Protocol (MCP) with a proper AI Gateway. This guide walks you through deploying MCP Servers on Kubernetes, adding authentication, locking down tools, and strengthening your organization’s overall MCP security posture.\ \ \ \ Read Blog](/content/blog/security-holes-in-mcp-servers-and-how-to-plug-them/index.html)

\ \ \ \ Announcing Gloo Mesh Support for Amazon ECS \ \ Latest Gloo Mesh release now provides support and enterprise-grade service mesh capabilities for Amazon ECS workloads.\ \ \ \ Read Blog](/content/blog/announcing-gloo-mesh-support-for-amazon-ecs/index.html)

\ \ \ \ Gloo Mesh 2.11: Expands Support to Amazon ECS and Brings Multi-Tenant Flexibility to Enterprises. \ \ Latest Gloo Mesh release expands support to Amazon ECS and brings multi-tenant flexibility to enterprises.\ \ \ \ Read Blog](/content/blog/gloo-mesh-2-11-release/index.html)

\ \ \ \ Reducing the costs and complexity of your cloud native architecture with Ambient Mesh \ \ Learn how Istio's Ambient Mesh simplifies cloud-native connectivity and dramatically reduces the cost and complexity of connecting, securing, and observing services across on-prem, cloud, or hybrid environments — without sidecars.\ \ \ \ Read Blog](/content/blog/ambient-mesh-reducing-cost-complexity-cloud-native-architecture/index.html)

\ \ \ \ Introducing Solo Enterprise for agentgateway \ \ From pilots to production with context-aware AI networking\ \ \ \ Read Blog](/content/blog/introducing-solo-enterprise-for-agentgateway/index.html)

\ \ \ \ Introducing Gloo Gateway 2.0 \ \ We're excited to introduce Gloo Gateway 2.0, built on the CNCF kgateway project and Kubernetes Gateway API. This release unifies open-source innovation with enterprise-grade extensions, ambient mesh integration, and AI-ready data planes to deliver secure, scalable, and future-proof API gateway capabilities for cloud-native and agentic workloads.\ \ \ \ Read Blog](/content/blog/introducing-gloo-gateway-2-0/index.html)

\ \ \ \ Ambient mesh deployments made easy with Gloo Operator \ \ This article discusses different ways to install Istio ambient mesh, and contrasts the Helm approach with the Gloo Operator, a new method for installing ambient mesh in Gloo Mesh.\ \ \ \ Read Blog](/content/blog/ambient-mesh-deployments-made-easy-with-gloo-operator/index.html)

\ \ \ \ Choosing between installation methods in Gloo Mesh: Helm vs. the Gloo Operator \ \ Explore Istio installation options with Gloo Mesh. Compare Helm for control and flexibility vs. Gloo Operator for simplicity and automation to find the best fit for your environment.\ \ \ \ Read Blog](/content/blog/gloo-mesh-installation-methods-helm-vs-gloo-operator/index.html)

\ \ \ \ How ambient mesh challenges the security gaps in sidecar workloads \ \ Discover how Istio’s ambient mesh strengthens microservices security beyond sidecars with improved isolation, reduced attack surfaces, and simpler operations.\ \ \ \ Read Blog](/content/blog/how-ambient-mesh-challenges-security-gaps-sidecar-workloads/index.html)

\ \ \ \ Migrating from sidecars to ambient with zero downtime \ \ Learn how to migrate from Istio sidecars to ambient mesh with zero downtime. Step-by-step strategies, best practices, and tools to ensure a safe transition.\ \ \ \ Read Blog](/content/blog/sidecars-ambient-zero-downtime/index.html)

\ \ \ \ Comparing Istio's ambient multicluster support with Gloo Mesh's multicluster peering \ \ Compare Istio’s new ambient multicluster support with Gloo Mesh’s multicluster peering. Learn the similarities, key differences, and scalability trade-offs.\ \ \ \ Read Blog](/content/blog/istio-ambient-multicluster-support-gloo-mesh-multicluster-peering/index.html)

\ \ \ \ The future of Kubernetes is context-aware: Meet Solo Enterprise for kagent \ \ Discover how Solo.io's enterprise version of kagent extends Kubernetes to turn cloud-native infrastructure into agent-native infrastructure.\ \ \ \ Read Blog](/content/blog/kagent-enterprise/index.html)

\ \ \ \ kgateway as Ingress for Ambient Service Mesh \ \ Explore how kgateway and Istio Ambient Mesh work together to deliver secure ingress, intelligent routing and clear observability.\ \ \ \ Read Blog](/content/blog/kgateway-ingress-ambient-service-mesh/index.html)

\ \ \ \ Tracing GenAI Applications Is Not Enough \ \ \ \ Read Blog](/content/blog/tracing-genai-applications-is-not-enough/index.html)

\ \ \ \ Gloo Mesh 2.10: More Secure, Scalable Cloud Connectivity \ \ Gloo Mesh 2.10 adds flat network support, traffic shifting, and policy enforcement for secure, scalable multi-cluster service mesh.\ \ \ \ Read Blog](/content/blog/gloo-mesh-secure-scalable-cloud-connectivity/index.html)

\ \ \ \ MCP Authorization is a Non-Starter for Enterprise\ \ In this blog, we highlight some of MCP's foundational challenges, alternative proposals in the community, and sharing our opinion on what this should look like in enterprise environments. We know the MCP community is hard at work on revising the specification and we feel future updates will align better with our recommendations here. \ \ \ \ Read Blog](/content/blog/mcp-authorization-is-a-non-starter-for-enterprise/index.html)

\ \ \ \ Securing and Observing Your Services, Simplified \ \ Istio Ambient Mesh’s ztunnel delivers secure-by-default microservices communication and real-time traffic visibility - without sidecars. Learn how it boosts performance, simplifies management, and reduces costs while enhancing security and observability.\ \ \ \ Read Blog](/content/blog/securing-observing-services-simplified/index.html)

\ \ \ \ From MCP Servers to Services: Introducing kmcp for Enterprise-Grade MCP Development \ \ \ \ Read Blog](/content/blog/introducing-kmcp/index.html)

\ \ \ \ The Power of a Single API to Secure, Observe, and Control Traffic in All Directions \ \ Learn how the Omni vision unifies traffic, security, and observability control across cloud-native systems with Gloo Mesh and Gloo Gateway.\ \ \ \ Read Blog](/content/blog/api-secure-observe-control-traffic/index.html)

%20a%20Bad%20Idea.png)\ \ \ \ Why Building Large Kubernetes Clusters Is (Still) a Bad Idea \ \ Running massive Kubernetes clusters might seem simpler, but it’s a trap. Learn why scaling a single cluster creates hidden performance, security, and reliability issues, and how Gloo Mesh with Ambient Mesh makes multi-cluster networking finally viable.\ \ \ \ Read Blog](/content/blog/why-building-large-kubernetes-clusters-is-still-a-bad-idea/index.html)

\ \ \ \ Fortifying Your Cloud Native Connectivity Security Posture with Solo and Ambient Mesh \ \ Strengthen your cloud-native security posture with Istio and ambient mesh. Learn how ambient enhances zero-trust architecture, simplifies mTLS, reduces attack surface, and decouples security from app logic, all with less operational overhead.\ \ \ \ Read Blog](/content/blog/cloud-native-connectivity-security-solo-ambient-mesh/index.html)

\ \ \ \ Migrating from Sidecars to Ambient Mesh - Risks, Challenges, and Benefits \ \ Considering migrating from Istio sidecars to ambient mesh? Learn about the key challenges, risks, and benefits of ambient, including improved performance, lower costs, and operational simplicity, plus tips to plan a safe, successful transition.\ \ \ \ Read Blog](/content/blog/ambient-mesh-migration/index.html)

\ \ \ \ Overhaul of Agent Gateway supporting A2A, MCP, and Kubernetes Gateway API \ \ Today, we’re excited to share the next major milestone: Agent Gateway is now a full-featured, AI-native gateway that combines deep MCP and A2A protocol awareness, robust traffic policy controls, inference gateway support, Kubernetes Gateway API support, and unified access to major LLMs, all purpose-built with Rust for real-world agentic systems.\ \ \ \ Read Blog](/content/blog/updated-a2a-and-mcp-gateway/index.html)

\ \ \ \ How Ambient Mesh Delivers Advanced Resource and Cost Savings \ \ Discover how Ambient Mesh architecture can reduce service mesh infrastructure costs by up to 92% compared to traditional sidecar deployments, with real-world savings\ \ \ \ Read Blog](/content/blog/how-ambient-mesh-delivers-advanced-resource-and-cost-savings/index.html)

\ \ \ \ Getting Started with Ambient Mesh: From 0 to 100 mph \ \ Learn how Ambient Mesh revolutionizes service mesh architecture by eliminating sidecars and introducing a split proxy approach for better performance and operational simplicity.\ \ \ \ Read Blog](/content/blog/getting-started-with-ambient-mesh-from-0-to-100-mph/index.html)

\ \ \ \ Agent Discovery, Naming, and Resolution - the Missing Pieces to A2A \ \ While the A2A specification provides the critical first steps toward discovery with Agent Cards, the infrastructure for truly dynamic, scalable agent ecosystems requires additional components that the spec intentionally leaves “up to you.” In this blog, we dig into those missing pieces. \ \ \ \ Read Blog](/content/blog/agent-discovery-naming-and-resolution---the-missing-pieces-to-a2a/index.html)

\ \ \ \ Part Two: MCP Authorization The Hard Way \ \ Digging into the details of the MCP Authorization Spec\ \ \ \ Read Blog](/content/blog/part-two-mcp-authorization-the-hard-way/index.html)

\ \ \ \ Part One: MCP Authorization The Hard Way \ \ Deep dive into MCP Authorization, step by step with examples and in-depth detail\ \ \ \ Read Blog](/content/blog/understanding-mcp-authorization-step-by-step-part-one/index.html)

\ \ \ \ Agent Identity and Access Management - Can SPIFFE Work? \ \ Digging into AI identity and how the current SPIFFE models may need to be revised to support AI Agents\ \ \ \ Read Blog](/content/blog/agent-identity-and-access-management---can-spiffe-work/index.html)

\ \ \ \ Deep Dive into llm-d and Distributed Inference \ \ Digging into the llm-d project and how it does distributed inference.\ \ \ \ Read Blog](/content/blog/deep-dive-into-llm-d-and-distributed-inference/index.html)

\ \ \ \ Gloo Mesh 2.8 simplifies service mesh operations with new enhanced user experience across multi-cluster environments. \ \ \ \ Read Blog](/content/blog/gloo-mesh-2-8-release/index.html)

\ \ \ \ Gloo Gateway 1.19 accelerates context-rich, real-time AI apps with Gateway API \ \ \ \ Read Blog](/content/blog/gloo-gateway-1-19-release/index.html)

\ \ \ \ llm-d: Distributed Inference Serving on Kubernetes \ \ \ \ Read Blog](/content/blog/llm-d-distributed-inference-serving-on-kubernetes/index.html)

\ \ Motive \ \ Motive modernized its infrastructure using Solo Enterprise for kgateway to boost reliability, developer agility, and fleet innovation.\ \ \ \ Read Case Study](/content/resources/case-study/motive/index.html)

\ \ Confluent \ \ Discover how Confluent achieved 100% mTLS coverage, FedRAMP-ready FIPS encryption, and real-time observability across 100+ services using Solo Enterprise for Istio.\ \ \ \ Read Case Study](/content/resources/case-study/confluent/index.html)

\ \ Ingenico \ \ Powered by Solo Enterprise for kgateway, Solo.io helped Ingenico modernize its global payments infrastructure—boosting scalability, resilience, and developer autonomy. Explore their journey to building a fault-tolerant, future-ready platform.\ \ \ \ Read Case Study](/content/resources/case-study/ingenico-2/index.html)

\ \ OfferUp \ \ OfferUp leveraged Kubernetes and Solo Enterprise for kgateway to modernize its marketplace, enabling developer self-service, faster deployments, and seamless scaling beyond peer-to-peer transactions.\ \ \ \ Read Case Study](/content/resources/case-study/offerup/index.html)

\ \ ParkMobile \ \ ParkMobile’s Platform Engineering team utilized Kubernetes and Solo Enterprise for kgateway to drive innovation, scalability, and seamless mobility solutions while fostering a culture of collaboration and technological excellence.\ \ \ \ Read Case Study](/content/resources/case-study/park-mobile/index.html)

\ \ Vonage \ \ Solo.io helped Vonage modernize its cloud infrastructure, enhancing scalability, reliability, and developer autonomy with Solo Enterprise for kgateway. Explore their journey to a building an efficient and agile platform.\ \ \ \ Read Case Study](/content/resources/case-study/vonage-2/index.html)

\ \ Domino’s Pizza \ \ Powered by Solo Enterprise for Istio and Solo Enterprise for kgateway, Solo.io helped Domino’s UK transition from a monolithic system to a microservices-based architecture. Learn about our 18-month journey to transform the way they operate.\ \ \ \ Read Case Study](/content/resources/case-study/dominos-pizza-uk/index.html)

\ \ Introducing Solo Enterprise for agentgateway \ \ Secure, govern, and operationalize AI agent connectivity at scale with Solo Enterprise for agentgateway\ \ \ \ Read Datasheet](/content/resources/datasheet/introducing-solo-enterprise-for-agentgateway/index.html)

\ \ Comparing Sidecars with Sidecarless Mesh Implementation \ \ Compare sidecar-based Istio with sidecarless Ambient Mesh. Learn how Gloo Mesh simplifies service mesh operations, reduces overhead, and enables scalable, secure, multi-cluster environments — with support for both migration paths.\ \ \ \ Read Datasheet](/content/resources/datasheet/comparing-sidecars-with-sidecarless-mesh-implementation/index.html)

\ \ Solo Enterprise for Istio Feature Comparison \ \ Compare features across Gloo Mesh Enterprise, Gloo Mesh Open Source, and Basic Open Source Istio.\ \ \ \ Read Datasheet](/content/resources/datasheet/gloo-mesh-feature-comparison/index.html)

\ \ Enterprise Support for Istio in Production \ \ Solo.io provides enterprise support for Istio environments to help you avoid pitfalls and resolve issues quickly.\ \ \ \ Read Datasheet](/content/resources/datasheet/enterprise-support-for-istio-in-production/index.html)

\ \ Service Mesh for Developers, Part 1: Exploring the Power of Observability and OpenTelemetry \ \ Navigating the complexity of modern applications requires a key ally – observability. Observability empowers teams to streamline application debugging, and within the architecture of a service mesh, provides valuable insights that increase reliability and performance.\ \ \ \ Read Ebook](/content/resources/ebook/service-mesh-for-developers-part-1/index.html)

\ \ Service Mesh at Scale \ \ Challenges and approaches to multi-cluster deployment patterns\ \ \ \ Read Ebook](/content/resources/ebook/service-mesh-at-scale/index.html)

\ \ Compare Capabilities of the Top Service Mesh Platforms \ \ Using a service mesh as a layer to unify communications between applications, services, and workloads empowers teams to modernize their systems, deliver faster results, and improve performance for modern enterprises.\ \ \ \ Read Ebook](/content/resources/ebook/buyers-guide-service-mesh/index.html)

\ \ Compare Capabilities of the Top API Gateways \ \ Download our Buyer’s Guide to API Gateways and learn about the modern requirements of API gateways in our guide’s comparison of five leading API gateway providers.\ \ \ \ Read Ebook](/content/resources/ebook/api-gateway-buyers-guide/index.html)

\ \ Establishing zero trust security for modern cloud architectures \ \ How your organization can ensure safer cloud architecture by applying a zero trust network security model\ \ \ \ Read Ebook](/content/resources/ebook/establishing-zero-trust-security-for-modern-cloud-architectures/index.html)

\ \ Unlocking the Power of Your API Gateway \ \ With automated API management, organizations save time and resources and streamline the development process.\ \ \ \ Read Ebook](/content/resources/ebook/unlocking-the-power-of-your-api-gateway/index.html)

\ \ API Gateways: Productivity, Resilience, and Security for Next-Generation Cloud Applications \ \ The rise of microservices architecture has brought about a significant shift in how software is developed and deployed, and as such, has presented new technical and organizational challenges.\ \ \ \ Read Ebook](/content/resources/ebook/api-gateways-productivity-resilience-and-security-for-next-generation-cloud-applications/index.html)

\ \ Driving Business Value with Istio \ \ How a service mesh can help your organization simplify the adoption of a distributed architecture\ \ \ \ Read Ebook](/content/resources/ebook/driving-business-value-with-istio/index.html)

\ \ Service Mesh Vendor Comparison \ \ See how top vendors with Istio-based service mesh offerings compare\ \ \ \ Read Ebook](/content/resources/ebook/istio-service-mesh-vendor-comparison/index.html)

\ \ Istio Then & Now \ \ \ \ Read Infographic](/content/resources/infographic/istio-then-now/index.html)

\ \ 4 Reasons Why You Need an AI Gateway \ \ Integrating LLM models in your applications? See our infographic to learn the top four reasons why you need an AI Gateway!\ \ \ \ Read Infographic](/content/resources/infographic/4-reasons-why-you-need-an-ai-gateway/index.html)

\ \ Solo Enterprise for kgateway vs. Kong \ \ Kong Gateway offers a very capable competitor to Gloo Gateway by Solo.io, but there are growing questions about the efficacy and stability of the Nginx open-source community behind the technology.\ \ \ \ Read Infographic](/content/resources/infographic/gloo-gateway-vs-kong/index.html)

\ \ Solo Enterprise for kgateway vs. Apigee \ \ Apigee suits Google Cloud, but its traditional approach clashes with modern practices. In contrast, Gloo Gateway by Solo.io aligns with cloud-native strategies.\ \ \ \ Read Infographic](/content/resources/infographic/gloo-gateway-vs-apigee/index.html)

\ \ 3 Reasons You Need an API Gateway for Microservices Apps \ \ Unlock the full potential of your microservices architecture with the strategic integration of API gateways. While microservices provide flexibility and scalability, they also introduce complexities that can impede seamless communication between services. Discover the crucial role of API gateways in overcoming these challenges and optimizing your microservices ecosystem.\ \ \ \ Read Infographic](/content/resources/infographic/3-reasons-you-need-an-api-gateway-for-microservices-apps/index.html)

\ \ Migrate to agentgateway with the ingress2gateway tool \ \ Learn to migrate your ingress-nginx Ingress configurations to agentgateway with the ingress2gateway tool. \ \ \ \ Read Lab](/content/resources/lab/migrate-to-agentgateway-with-the-ingress2gateway-tool/index.html)

\ \ Migrate to kgateway with the ingress2gateway tool \ \ Learn to migrate your ingress-nginx Ingress configurations to kgateway with the ingress2gateway tool.\ \ \ \ Read Lab](/content/resources/lab/migrate-to-kgateway-with-the-ingress2gateway-tool/index.html)

\ \ Introduction to agentregistry \ \ Free Lab: Learn how to curate, publish, and deploy MCP servers and AI skills using agentregistry for unified management of AI-native artifacts.\ \ \ \ Read Lab](/content/resources/lab/introduction-to-agentregistry/index.html)

\ \ Build AI agents with agent skills \ \ Free Lab: Learn to build a declarative AI agent in kagent with a Kubernetes deployment skill, enabling automated app deployments and MCP tool integration.\ \ \ \ Read Lab](/content/resources/lab/build-ai-agents-with-agent-skills/index.html)

\ \ Program agentgateway for LLM consumption \ \ Free Lab: Learn to proxy OpenAI requests through an agentgateway-backed AI gateway with kgateway, managing LLM traffic, credentials, and advanced features.\ \ \ \ Read Lab](/content/resources/lab/program-agentgateway-for-llm-consumption/index.html)

\ \ Local development with the kagent CLI \ \ Free Lab: Learn to build, run, and deploy AI agents with kagent, integrate MCP servers, and manage tools in Kubernetes for AI-native workloads.\ \ \ \ Read Lab](/content/resources/lab/local-development-with-the-kagent-cli/index.html)

\ \ Secure your MCP servers with OAuth \ \ Free Lab: Learn to build, proxy, and secure an MCP server with agentgateway and OAuth2, protecting AI-native workloads while enabling safe tool access in Kubernetes.\ \ \ \ Read Lab](/content/resources/lab/secure-your-mcp-servers-with-oauth/index.html)

\ \ Improve cloud native operations & troubleshooting with kagent \ \ Free Lab: Use AI agents with kgateway and kagent to automate Kubernetes tasks, provision gateways, route traffic, and manage workloads in a cloud-native environment.\ \ \ \ Read Lab](/content/resources/lab/improve-cloud-native-ops-troubleshooting/index.html)

\ \ Observe Agent & MCP server interactions \ \ Free Lab: Learn how to use Solo Enterprise for kagent to create, run, and monitor AI agents in Kubernetes with full observability and control.\ \ \ \ Read Lab](/content/resources/lab/observe-agent-mcp-server-interactions/index.html)

\ \ Multiplex MCP servers & control auth policy \ \ Solo Free Lab: Learn how to use kgateway and agentgateway to deploy, proxy, and secure MCP servers in Kubernetes with AI-native control and traffic routing.\ \ \ \ Read Lab](/content/resources/lab/multiplex-mcp-servers-control-auth-policy/index.html)

\ \ Build, run & deploy MCP servers to Kubernetes \ \ Solo Free Lab: Learn to use the new kmcp tool to easily deploy MCP servers to Kubernetes.\ \ \ \ Read Lab](/content/resources/lab/build-run-deploy-mcp-servers-to-kubernetes/index.html)

\ \ Kagent Lab: Discover kagent and kmcp \ \ Explore the kagent project to build custom agents and tools\ \ \ \ Read Lab](/content/resources/lab/kagent-lab-discover-kagent-kmcp/index.html)

\ \ Solo Enterprise for Istio Lab: OpenTelemetry collectors and relay \ \ Solo.io Free Lab: Learn about the relay of telemetry from a workload cluster to the management cluster in Gloo Mesh.\ \ \ \ Read Lab](/content/resources/lab/gloo-mesh-otel-collectors-relay/index.html)

\ \ Solo Enterprise for Istio Lab: Extended telemetry from ztunnel \ \ Solo.io Free Lab: Learn how to deploy Istio Ambient Mesh with Gloo Mesh to capture Layer 7 telemetry directly from zTunnel - no waypoints required.\ \ \ \ Read Lab](/content/resources/lab/gloo-mesh-extended-telemetry-ztunnel/index.html)

\ \ Solo Enterprise for Istio Lab: Configure enhanced waypoint proxies \ \ Solo.io Free Lab: Learn how to use Gloo Gateway for your waypoints in Gloo Mesh, instead of Istio's default waypoint proxy.\ \ \ \ Read Lab](/content/resources/lab/gloo-mesh-lab-enhanced-waypoint-proxies/index.html)

\ \ Solo Enterprise for Istio Lab: Multicluster peering \ \ Solo.io Free Lab: Learn how to enable multicluster peering in Gloo Mesh to run services across clusters as a single mesh, making workloads visible and accessible between clusters with global failover and redundancy.\ \ \ \ Read Lab](/content/resources/lab/gloo-mesh-lab-multicluster-peering/index.html)

\ \ Ambient Mesh Lab: EnvoyFilter Support \ \ Solo.io Free Lab: Learn how to preserve and migrate EnvoyFilter configurations when transitioning from sidecar to ambient mode in Istio using Solo.io’s extended build, with step-by-step guidance and validation.\ \ \ \ Read Lab](/content/resources/lab/ambient-mesh-lab-envoyfilter-support/index.html)

\ \ Ambient Mesh Lab: SPIRE integration with Gloo Mesh in Istio Ambient Mode \ \ Secure your Istio Ambient Mesh with SPIRE. This hands-on lab shows how to integrate SPIRE with Gloo Mesh to issue SPIFFE identities and certificates for ztunnel and mesh workloads.\ \ \ \ Read Lab](/content/resources/lab/gloo-mesh-lab-spire-integration-with-gloo-mesh-in-ambient-mode/index.html)

\ \ Ambient Mesh Lab: Introduction to ztunnel in Ambient Mesh \ \ Learn how Ambient Mesh uses ztunnel to secure traffic without sidecars. This free lab walks you through joining workloads, observing traffic, verifying mTLS, and applying Layer 4 policies.\ \ \ \ Read Lab](/content/resources/lab/gloo-mesh-lab-intro-ztunnel-ambient-mesh/index.html)

\ \ Solo Academy Course: Service Mesh Basics \ \ Solo Academy | Learn the fundamentals of service mesh in this free video-led course and get insights into using a service mesh to enhance your observability, security and reliability of your applications \ \ \ \ Read Lab](/content/resources/lab/solo-academy-course-service-mesh-basics/index.html)

\ \ Solo Academy Course: Istio Basics \ \ Solo Academy | Learn the basics of Istio with our free 101 course. Understand what Istio is, how it works, and its features for traffic management, security, and observability in microservices and Kubernetes.\ \ \ \ Read Lab](/content/resources/lab/solo-academy-course-istio-basics/index.html)

\ \ Solo Academy Course: Envoy Basics \ \ Solo Academy | Master Envoy Proxy basics with our free 101 level course. Learn about the architecture, advanced load balancing, observability, and role in microservices, Kubernetes, and service meshes\ \ \ \ Read Lab](/content/resources/lab/solo-academy-course-envoy-basics/index.html)

\ \ Solo Academy Course: API Gateway Basics \ \ Solo Academy | Learn API Gateway fundamentals in Kubernetes with this free beginner level video-led course. Discover how API Gateways control traffic, secure connectivity, and complement microservices architecture\ \ \ \ Read Lab](/content/resources/lab/solo-academy-course-api-gateway-basics/index.html)

\ \ Solo Academy Course: Get Started with Istio Service Mesh \ \ Solo Academy | A fundamental level workshop for developers and operators to learn Istio service mesh. Install Istio, secure services, control traffic, and earn a Solo.io certification through hands-on labs.\ \ \ \ Read Lab](/content/resources/lab/solo-academy-course-get-started-with-istio-service-mesh/index.html)

\ \ Solo Academy Course: Deploying Istio for Production \ \ Solo Academy | Free hands-on workshop for operators looking to deploy Istio in production. Learn advanced routing, observability, security, mTLS, and multi-cluster setup and g free certification. \ \ \ \ Read Lab](/content/resources/lab/solo-academy-course-deploying-istio-for-production/index.html)

\ \ Kgateway Lab: Integrating kgateway with Istio at Ingress \ \ Explore how to integrate kgateway's ingress gateway with Istio Ambient Mesh in this free hands-on lab. Learn to deploy workloads, configure Gateway and Route resources, and enable automatic mutual TLS between the gateway and backend services.\ \ \ \ Read Lab](/content/resources/lab/kgateway-lab-integrating-kgateway-with-istio-at-ingress/index.html)

\ \ Kgateway Lab: Kgateway as a Waypoint \ \ Learn how to deploy and configure kgateway as a waypoint in Istio Ambient Mesh. This free hands-on lab walks you through managing east-west traffic, applying custom policies, and enhancing service communication with enterprise-grade control.\ \ \ \ Read Lab](/content/resources/lab/kgateway-lab-kgateway-as-a-waypoint/index.html)

\ \ Kgateway AI Lab: Deploying kgateway as an AI Gateway \ \ Learn how to enable the AI extension, configure gateway parameters, and deploy an AI Gateway using kgateway to route requests to large language models (LLMs) from within your Kubernetes cluster.\ \ \ \ Read Lab](/content/resources/lab/deploying-kgateway-as-an-ai-gateway/index.html)

\ \ Kagent Lab: How to build an AI agent \ \ Create Your First AI Agent with Kagent in Kubernetes\ \ \ \ Read Lab](/content/resources/lab/kagent-lab-how-to-build-an-ai-agent/index.html)

\ \ Kagent Lab: Integrate tools from MCP servers with kagent \ \ Kagent Lab: Integrate tools from MCP servers with kagent\ \ \ \ Read Lab](/content/resources/lab/kagent-lab-integrate-tools-from-mcp-servers-with-kagent/index.html)

\ \ agentgateway Hands-On Lab: Semantic Caching \ \ agentgateway Labs | Semantic Caching with Gloo AI Gateway \ \ \ \ Read Lab](/content/resources/lab/gloo-ai-gateway-hands-on-lab-semantic-caching/index.html)

\ \ Kgateway AI Lab: Credentials Management \ \ kgateway labs | Managing LLM Credentials with kgateway - AI Gateway\ \ \ \ Read Lab](/content/resources/lab/kgateway-ai-lab-credentials-management/index.html)

\ \ Kgateway AI Lab: Prompt Enrichment \ \ Kgateway labs | Managing Prompts for Enhanced LLM Performance\ \ \ \ Read Lab](/content/resources/lab/kgateway-ai-lab-prompt-enrichment/index.html)

\ \ Kgateway AI Lab: Prompt Guards \ \ kgateway labs | Content Safety with Prompt Guards \ \ \ \ Read Lab](/content/resources/lab/kgateway-ai-lab-prompt-guards/index.html)

\ \ Ambient Mesh Lab: Migrating from Sidecar to Sidecarless \ \ Ambient Mesh Lab | Migrate from Sidecar-based Service Mesh to Ambient Mesh \ \ \ \ Read Lab](/content/resources/lab/ambient-mesh-lab-migrating-from-sidecar-to-sidecarless/index.html)

\ \ Ambient Mesh Lab: Multi-cluster scalability with Istio Ambient Mesh \ \ Mastering Multi-Cluster Scalability with Ambient Mesh\ \ \ \ Read Lab](/content/resources/lab/ambient-mesh-lab-multi-cluster-scalability-with-istio-ambient-mesh/index.html)

\ \ Solo Lab: Gloo Cloud Preview \ \ Simplify Mesh Management with Gloo Cloud: Onboarding, Ingress, Egress, and Service Mesh\ \ \ \ Read Lab](/content/resources/lab/solo-lab-gloo-cloud-preview/index.html)

\ \ Ambient Mesh Lab: Waypoints for Traffic management, Security and Observability \ \ Solo Lab | Waypoints in Ambient Mesh: For L4 and L7 Security, Traffic and Observability Insights\ \ \ \ Read Lab](/content/resources/lab/ambient-mesh-lab-waypoints-for-traffic-management-security-and-observability/index.html)

\ \ Kgateway Lab: Gateway API inference extensions with kgateway \ \ Exploring the Gateway API Inference Extension with kgateway\ \ \ \ Read Lab](/content/resources/lab/kgateway-lab-gateway-api-inference-extensions-with-kgateway/index.html)

\ \ Kgateway Lab: Securing access to workloads with Gloo Gateway \ \ Securing Services with Gloo Gateway: TLS Termination and API Keys\ \ \ \ Read Lab](/content/resources/lab/gloo-gateway-lab-securing-access-to-workloads-with-gloo-gateway/index.html)

\ \ Kgateway Lab: Route Delegation in kgateway \ \ Route Delegation in kgateway\ \ \ \ Read Lab](/content/resources/lab/route-delegation-in-kgateway/index.html)

\ \ Kgateway Lab: Canary releases with Argo Rollouts & kgateway \ \ Canary releases with Argo Rollouts & kgateway\ \ \ \ Read Lab](/content/resources/lab/canary-releases-with-argo-rollouts-kgateway/index.html)

\ \ Kgateway Lab: Understanding kgateway and Gateway API policy attachments \ \ Understanding kgateway patterns of extensions \ \ \ \ Read Lab](/content/resources/lab/understanding-kgateway-patterns-of-extensions/index.html)

\ \ Kgateway Lab: Gateway API support for service mesh with kgateway \ \ GatewayAPI support for service mesh with kgateway\ \ \ \ Read Lab](/content/resources/lab/gatewayapi-support-for-service-mesh-with-kgateway/index.html)

\ \ Kgateway Lab: Exploring HTTPRoute resource configurations with kgateway \ \ Exploring HTTPRoute resource configurations with kgateway\ \ \ \ Read Lab](/content/resources/lab/exploring-httproute-resource-configurations-with-kgateway/index.html)

\ \ Kgateway Lab: Configuring gateways across multiple teams with kgateway\ \ Configuring gateways across multiple teams with kgateway\ \ \ \ Read Lab](/content/resources/lab/configuring-gateways-across-multiple-teams-with-kgateway/index.html)

\ \ Kgateway Lab: Configure HTTPS with the Gateway API and kgateway \ \ Configure HTTPS with the Gateway API and kgateway\ \ \ \ Read Lab](/content/resources/lab/configure-https-with-the-gateway-api-and-kgateway/index.html)

\ \ Kgateway Lab: Understanding the basics of Kubernetes Gateway API with kgateway \ \ Understanding the basics of Kubernetes Gateway API with kgateway\ \ \ \ Read Lab](/content/resources/lab/understanding-the-basics-of-kubernetes-gateway-api-with-kgateway/index.html)

\ \ Kgateway Lab: Installing kgateway, an open-source implementation of the Kubernetes Gateway API \ \ Installing kgateway, an open-source implementation of the Kubernetes Gateway API\ \ \ \ Read Lab](/content/resources/lab/install-kgateway-open-source-implementation-of-the-gateway-api/index.html)

\ \ Ambient Mesh Lab: Employing circuit breaking in Ambient Mesh \ \ Join our free, on-demand lab Employing Circuit Breaking to safeguard services with Istio Ambient mode. Learn to deploy waypoints, configure circuit breaking, and monitor using metrics, logs, and Prometheus.\ \ \ \ Read Lab](/content/resources/lab/employing-circuit-breaking-in-ambient-mesh/index.html)

\ \ Ambient Mesh Lab: Configuring Fault Injection in Ambient Mesh \ \ Join our free, on-demand lab Configuring Fault Injection to enhance resiliency with Istio Ambient mode. Learn to observe system latency, configure delays, timeouts, retries, and augment with outlier detection.\ \ \ \ Read Lab](/content/resources/lab/configuring-fault-injection-in-ambient-mesh/index.html)

\ \ Ambient Mesh Lab: Using Outlier Detection with Ambient Mesh \ \ Join our free, on-demand lab Using Outlier Detection to learn how to configure Istio Ambient mode to avoid unhealthy workloads. Discover steps to implement outlier detection, assess workload health, and monitor metrics effectively.\ \ \ \ Read Lab](/content/resources/lab/using-outlier-detection-with-ambient-mesh/index.html)

\ \ Ambient Mesh Lab: Implementing Timeouts with Ambient Mesh \ \ Join our free on-demand lab, Implementing Timeouts, and learn how to protect applications from slow upstream services with Istio. Discover how to prevent indefinite errors, configure and verify timeouts, and decouple resiliency concerns from your apps.\ \ \ \ Read Lab](/content/resources/lab/implementing-timeouts/index.html)

\ \ Kgateway Lab: Exposing, Securing, and Rate Limiting with kgateway \ \ Free K8s Gateway Lab: Deploy, Secure, and Rate Limiting with Solo's Open Source Gateway\ \ \ \ Read Lab](/content/resources/lab/exposing-securing-and-rate-limiting-with-k8s-gateway/index.html)

\ \ Ambient Mesh Lab: Traffic routing with waypoints in Ambient Mesh \ \ Learn how to route traffic using waypoint proxies in ambient mesh.\ \ \ \ Read Lab](/content/resources/lab/traffic-routing-with-waypoints/index.html)

\ \ Ambient Mesh Lab: Secure Your Cluster with Ambient Mesh and mTLS \ \ Learn how to secure services in your Kubernetes cluster using ambient mesh and mTLS.\ \ \ \ Read Lab](/content/resources/lab/secure-your-cluster-with-ambient-mesh-and-mtls/index.html)

\ \ Ambient Mesh Lab: Access control with authorization policies \ \ Learn how to enforce access control and write authorization policies for L4 and L7 traffic.\ \ \ \ Read Lab](/content/resources/lab/access-control-with-authorization-policies/index.html)

\ \ Ambient Mesh Lab: Getting Started with Ambient Mesh \ \ Learn the basics of ambient and how to set up your first environment.\ \ \ \ Read Lab](/content/resources/lab/getting-started-with-ambient-mesh/index.html)

\ \ Ambient Mesh Lab: Getting L4 and L7 observability \ \ Learn how to view metrics and traces from L4 and L7 traffic in ambient mesh.\ \ \ \ Read Lab](/content/resources/lab/getting-l4-and-l7-observability/index.html)

\ \ agentgateway Hands-On Lab: Prompt Management and Prompt Guards \ \ Sign up for the free, hands-on technical labs.\ \ \ \ Read Lab](/content/resources/lab/gloo-ai-gateway-hands-prompt-management-prompt-guards/index.html)

\ \ agentgateway Hands-On Lab: Rate Limiting and Model Failover \ \ Sign up for the free, hands-on technical labs.\ \ \ \ Read Lab](/content/resources/lab/gloo-ai-gateway-rate-limiting-model-failover/index.html)

\ \ agentgateway Hands-On Lab: RAG and Semantic Caching \ \ Sign up for the free, hands-on technical labs.\ \ \ \ Read Lab](/content/resources/lab/gloo-ai-gateway-rag-semantic-caching/index.html)

\ \ agentgateway Hands-On Lab: Credentials and Access Control \ \ Sign up for the free, hands-on technical labs.\ \ \ \ Read Lab](/content/resources/lab/gloo-ai-gateway-credentials-access-control/index.html)

\ \ AI Agents in Kubernetes \ \ \ \ Read Report](/content/resources/report/ai-agents-in-kubernetes/index.html)

\ \ AI Gateways in the Enterprise \ \ \ \ Read Report](/content/resources/report/ai-gateways-in-the-enterprise/index.html)

\ \ API Gateway Resource Kit \ \ A service mesh is a dedicated infrastructure layer that helps manage and secure communications between microservices within a distributed application.\ \ \ \ Read Blog](/content/resources/resource-kit/api-gateway/index.html)

\ \ Service Mesh Resource Kit \ \ A service mesh is a dedicated infrastructure layer that helps manage and secure communications between microservices within a distributed application.\ \ \ \ Read Blog](/content/resources/resource-kit/service-mesh/index.html)

\ \ AI Agents Are Not APIs. Existing Gateways Can't Tell the Difference. \ \ \ \ Read Whitepaper](/content/resources/white-paper/ai-agents-are-not-apis/index.html)

\ \ Building Responsive and Resilient Multi-Cluster Applications with Solo’s Ambient Mesh \ \ \ \ Read Whitepaper](/content/resources/white-paper/building-responsive-and-resilient-multi-cluster-applications-with-solos-ambient-mesh/index.html)

\ \ Busting 10 Myths About Istio Ambient and Sidecarless Service Mesh \ \ \ \ Read Whitepaper](/content/resources/white-paper/busting-10-myths-about-istio-ambient/index.html)

\ \ Guide to Migrating from Ingress to Gateway API \ \ Migrate your project from ingress-nginx and the Ingress API to the Kubernetes Gateway API with kgateway—an open-source Gateway project powered by Envoy.\ \ \ \ Read Whitepaper](/content/resources/white-paper/guide-to-migrating-from-ingress-to-gateway-api/index.html)

\ \ Migrating from Sidecars to Sidecarless Istio Ambient Mesh \ \ In our white paper Migrating from Sidecars to Sidecarless to Ambient Mesh, we share how sidecars and Ambient mode can work together, allowing for a gradual migration strategy.\ \ \ \ Read Whitepaper](/content/resources/white-paper/migrating-from-sidecars-to-sidecarless-istio-ambient-mesh/index.html)

\ \ Introduction and Best Practices to AI Gateways \ \ \ \ Read Whitepaper](/content/resources/white-paper/intro-best-practices-to-ai-gateways/index.html)

\ \ Solo.io’s Guide to Navigating GenAI Complexity \ \ \ \ Read Whitepaper](/content/resources/white-paper/guide-to-navigating-genai-complexity/index.html)

\ \ Unlocking Business Efficiency with Service Mesh Updates in AWS \ \ \ \ Read Whitepaper](/content/resources/white-paper/service-mesh-aws/index.html)

\ \ Evolve Your API Management \ \ \ \ Read Whitepaper](/content/resources/white-paper/evolve-your-api-management/index.html)

\ \ Transitioning From App Mesh to Istio for AWS EKS \ \ \ \ Read Whitepaper](/content/resources/white-paper/app-mesh-istio-aws-eks-transition/index.html)

\ \ How Service Mesh Supports a Zero Trust Architecture \ \ \ \ Read Whitepaper](/content/resources/white-paper/how-service-mesh-supports-a-zero-trust-architecture/index.html)

\ \ Achieve Compliance, Zero Trust with Istio Ambient Mode \ \ \ \ Read Whitepaper](/content/resources/white-paper/achieve-compliance-zero-trust-with-istio-ambient-mesh/index.html)

\ \ Get started \ \ See why Solo.io is the leading provider of API Gateway and service mesh solutions.\ \ \ \ Learn more](/content/get-started/index.html)

\ \ Istio support \ \ Is Istio not working properly for you? Are you having performance issues with your Istio instance? Is there an upgrade you need to manage on multiple clusters?\ \ \ \ Get support](/content/istio-support/index.html)

\ \ Pricing \ \ Reach out for tailored pricing options and step into a future of enhanced connectivity.\ \ \ \ Get Started](/content/pricing/index.html)

\ \ Book a Gloo AI Gateway demo \ \ Powered by Istio, Gloo Mesh empowers platform engineering teams to boost security, resiliency, and observability.\ \ \ \ Get started](/content/products/gloo-ai-gateway/demo/index.html)

\ \ Book a Gloo Gateway demo \ \ Gloo Gateway is a fast, Kubernetes- native API gateway packed with features IT operations teams need to deliver a full lifecycle security strategy for their cloud-native environments.\ \ \ \ Get started](/content/products/gloo-gateway/demo/index.html)

\ \ Gloo Mesh - Ambient Readiness Assessment \ \ A free program designed to assess your readiness for Istio Ambient adoption and prove the business benefits based on your own environment.\ \ \ \ Get started](/content/products/gloo-mesh/ambient-readiness-assessment/index.html)

\ \ Book a Gloo Mesh demo \ \ A free program designed to assess your readiness for Istio Ambient adoption and prove the business benefits based on your own environment.\ \ \ \ Get started](/content/products/gloo-mesh/demo/index.html)

\ \ Case Studies \ \ \ \ Learn more](/content/resources/case-study/index.html)

\ \ Ebooks \ \ \ \ Learn more](/content/resources/ebook/index.html)

\ \ Labs \ \ \ \ Learn more](/content/resources/lab/index.html)

\ \ Reports \ \ \ \ Read Reports](/content/resources/report/index.html)

\ \ Resource Kits \ \ \ \ See Resource Kits](/content/resources/resource-kit/index.html)

\ \ Webinars \ \ A free program designed to assess your readiness for Istio Ambient adoption and prove the business benefits based on your own environment.\ \ \ \ Watch Webinars](/content/resources/webinar/index.html)

\ \ Whitepapers \ \ A free program designed to assess your readiness for Istio Ambient adoption and prove the business benefits based on your own environment.\ \ \ \ Read Whitepapers](/content/resources/white-paper/index.html)

\ \ Gloo AI Gateway \ \ Secure, observe, and control your AI applications with Gloo AI Gateway, the leading cloud native gateway built on Envoy.\ \ \ \ Learn more](/content/dev/backup/gloo-ai-gateway/index.html)

\ \ Gloo Gateway \ \ The future of cloud APIs is omni.\ \ \ \ Learn more](/content/products/kgateway/index.html)

\ \ Gloo Mesh \ \ Connecting, securing, controlling, and observing microservice communication is tough. We make it accessible to all.\ \ \ \ Learn more](/content/products/istio/index.html)

\ \ Datasheets \ \ \ \ Read Datasheets](/content/resources/datasheet/index.html)

\ \ Infographics \ \ \ \ Read Infographics](/content/resources/infographic/index.html)

\ \ Videos \ \ \ \ Watch Videos](/content/resources/video/index.html)

Cloud connectivity done right

Get started

Native capabilities and the reference implementation

Why AI spend needs gateway control

Runaway agent loops

Context growth

Multi-model sprawl

Blunt provider limits

A concrete customer scenario

Architecture

How enforcement works

Why reservation and settlement are both necessary

The enforcement decision belongs in RequestHeaders

ExtProc phases

RequestHeaders

RequestBody

ResponseHeaders

ResponseBody

Calculating cost

Treat pricing as versioned configuration

Identity and the trust boundary

Budget model

Isolation and fallback

Periods and retry behavior

Approval workflow

Admin roles and permissions

Model pricing and unknown models

Use a safe unknown-model policy

Production hardening

Security guidance

Operating the system

Alerts and forecasting

Budget detail and management UI

Observability

Complement budgets with rate limits and token caps

Minimum adaptation path

1. Deploy the data store

2. Deploy the budget ExtProc service

3. Validate identity at the gateway

4. Attach ExtProc with a fail-closed policy

5. Seed model prices

6. Create initial budgets

7. Test allowed traffic

8. Test rejection behavior

9. Test failure modes

Example: Claude Code through the gateway

Validate retry behavior for your client version

Optimization after enforcement

Route by task shape

Shadow cheaper models

Cache repeat work

Track model usage

Summary

Learn more about agentgateway

Featured content

Cloud connectivity done right

The enforcement decision belongs in `RequestHeaders`

`RequestHeaders`

`RequestBody`

`ResponseHeaders`

`ResponseBody`