Beiträge

Veröffentlicht am

The Agent Egress Boundary: Making Every AI Tool Call Enforceable and Observable

AI agents do not create risk only when they generate the wrong answer. They create operational risk when they turn that answer into an outbound action: calling an API, querying a search service, downloading content, opening a ticket, sending a message, or changing a production system.

Most enterprise controls still focus on the agent’s intent. Prompts, guardrails, and model policies describe what the agent should do. They do not guarantee which destinations the workload can reach, which request was sent, or whether an unapproved path was used.

That gap calls for an agent egress boundary: a platform-enforced control through which every external tool call must pass, combined with traceable evidence that links the call to the originating agent interaction.

Guardrails are necessary, but they are not enforcement

Prompt-level guardrails are useful for shaping behavior. They can tell an agent not to disclose sensitive information, not to call unknown services, or to request human approval before a consequential action. But those controls operate inside the reasoning path they are intended to constrain.

Production systems need an independent layer. If an agent is compromised through prompt injection, a poisoned tool response, a vulnerable dependency, or a simple implementation mistake, the network should still prevent access to destinations outside the approved contract.

The distinction is familiar from other areas of security:

  • application authorization expresses intended access;
  • network enforcement limits reachable destinations;
  • observability records what actually happened;
  • human approval controls high-impact exceptions.

No single layer is sufficient. Together, they create defense in depth.

The platform contract

An agent egress boundary should answer four questions for every outbound request:

  1. Who initiated it? Identify the workload, agent, tenant, and user or workflow context.
  2. Where is it going? Resolve the approved destination, protocol, port, and application-level route.
  3. Was it allowed? Evaluate the call against a versioned policy rather than an application convention.
  4. What evidence remains? Record a traceable decision without leaking secrets or sensitive payloads.

This turns outbound connectivity into a platform contract. An agent receives only the network access required by its tools, while the platform provides a consistent control and evidence plane.

A practical cloud-native pattern

A recent CNCF implementation demonstrates the core idea using NGINX, Kubernetes, and OpenTelemetry. NGINX acts as both the inbound reverse proxy and the outbound forward proxy for an agent workload. Network rules drop direct egress so the proxy becomes the only approved path. The NGINX OpenTelemetry module emits a span for each request, and an OpenTelemetry Collector forwards the evidence to observability or security systems.

The important principle is architectural: the boundary is not a library the agent may choose to call. It is the only network path available.

A production-oriented request flow can look like this:

  1. A user or system invokes the agent through an authenticated gateway.
  2. The gateway propagates a trace context and workload identity.
  3. The agent selects a tool and issues an outbound request.
  4. Kubernetes egress controls permit traffic only to the designated proxy.
  5. The proxy evaluates destination, protocol, identity, and policy.
  6. Allowed traffic is forwarded; denied traffic returns a controlled error.
  7. OpenTelemetry records the decision and correlates it with the originating interaction.

The result is a chain of evidence from user request to external side effect.

Why Kubernetes NetworkPolicy alone is not enough

Kubernetes NetworkPolicy is a strong foundation. It can isolate workloads and restrict egress by IP block, port, and selected peers, provided the cluster’s network plugin enforces the policy. A default-deny egress policy should be the starting point for sensitive agent workloads.

However, many agent tools call dynamic external services over HTTPS. IP addresses change, destinations share infrastructure, and business rules are usually expressed in terms of domains, API routes, methods, or tool identities rather than static addresses.

That is why a layered design is useful:

  • NetworkPolicy or equivalent CNI controls ensure the workload can only reach the approved proxy and essential platform services.
  • The egress proxy enforces destination and application-aware rules.
  • Workload identity distinguishes agents and tenants without relying only on source IP.
  • OpenTelemetry provides correlated evidence for operations, security, and audit.

The network layer prevents bypass. The proxy layer understands enough context to make a useful decision.

Policy should follow the tool contract

Allowing an agent to reach an entire domain is often broader than the tool definition requires. A better policy starts with the declared tool contract.

For example, an incident-analysis agent may need to:

  • read selected observability APIs;
  • create, but not delete, incident tickets;
  • query a controlled knowledge source;
  • send notifications only to an approved channel;
  • never call arbitrary internet destinations.

The platform can translate that contract into an egress policy covering destination, method, route, identity, rate, and approval requirements. High-risk actions can be routed through a separate approval service rather than granted as normal network access.

This also creates a cleaner ownership model. Domain teams define which tools are necessary. Security teams define control requirements. Platform teams provide the reusable enforcement mechanism.

Observability must produce evidence, not surveillance

OpenTelemetry is well suited to correlating inbound interactions with outbound HTTP client activity. Standard HTTP span conventions provide consistent attributes for requests and responses, while trace context links multiple services into one transaction.

But recording everything is not automatically safe. Agent traffic can include credentials, personal data, customer information, prompts, and tool payloads. The audit plane therefore needs its own policy.

Useful evidence

  • trace and request identifiers;
  • agent, workload, tenant, and tool identity;
  • policy version and allow or deny decision;
  • destination service and approved route classification;
  • HTTP method and status class;
  • latency, retries, and byte counts;
  • model or agent configuration version;
  • human approval reference where required.

Data to avoid by default

  • authorization headers and API keys;
  • full request or response bodies;
  • raw prompts containing confidential data;
  • URL query parameters unless explicitly sanitized;
  • unbounded high-cardinality attributes.

The purpose is to prove and investigate behavior, not to create a second uncontrolled copy of sensitive data.

Controls that make the boundary credible

A proxy is only a boundary when bypass is demonstrably difficult. Platform teams should validate at least the following controls:

  • Default-deny egress: direct external connectivity fails.
  • DNS control: workloads cannot switch to an unmonitored resolver or exploit unexpected resolution paths.
  • IPv4 and IPv6 parity: policy applies consistently to both address families.
  • Protocol coverage: non-HTTP tools, WebSockets, streaming APIs, and message protocols have explicit handling.
  • TLS design: the organization decides where TLS terminates and what metadata can be inspected without undermining privacy.
  • Identity: decisions rely on authenticated workload identity, not only mutable labels or network location.
  • Fail-closed behavior: proxy, collector, or policy failures do not silently open direct access.
  • High availability: the control plane does not become an avoidable single point of failure.

These details determine whether the pattern is an architectural control or merely a useful demonstration.

Operational signals for platform teams

Once all tool traffic crosses the boundary, the same telemetry can improve reliability and cost control.

Useful service-level indicators include:

  • allowed and denied tool calls by agent and policy version;
  • unexpected destinations or repeated policy violations;
  • external dependency latency and error rates;
  • retry storms and rate-limit responses;
  • egress volume and estimated third-party API cost;
  • calls that required human approval;
  • trace gaps where an outbound action lacks an originating interaction.

This gives security and operations teams a shared view. The same denied request may indicate an attack, an outdated policy, or a legitimate new tool requirement.

A phased adoption plan

  1. Inventory agent egress. Identify destinations, protocols, credentials, and business owners for each production tool.
  2. Introduce observation first. Capture sanitized outbound traces to understand real behavior before enforcing a narrow policy.
  3. Define tool-level contracts. Document approved destinations and actions rather than granting general internet access.
  4. Apply default deny. Force a low-risk agent through the proxy and prove that direct egress fails.
  5. Add policy-as-code. Version destination rules, ownership, exceptions, and approval conditions in Git.
  6. Connect the audit plane. Send sanitized OpenTelemetry data to the organization’s observability and SIEM platforms.
  7. Test failure modes. Validate DNS bypass, IPv6, proxy outage, collector outage, policy rollback, and certificate rotation.
  8. Scale by platform product. Offer the boundary as a reusable golden-path capability rather than a custom design for every agent.

Conclusion

Enterprises should not have to trust that an AI agent will respect its network boundaries. Those boundaries should be enforced by the platform and evidenced through telemetry.

NGINX, Kubernetes, and OpenTelemetry show that the core pattern can be built from mature cloud-native components: default-deny connectivity, an application-aware egress proxy, and correlated traces. The exact implementation will vary, but the platform contract should remain consistent.

Every agent tool call should be attributable, policy-checked, observable, and reversible where the downstream system allows it. That is the difference between experimenting with autonomous software and operating it responsibly.

Sources and further reading

Veröffentlicht am

The AI-Native Platform Contract: Expanding Golden Paths Beyond Application Delivery

Platform engineering earned its place by turning application delivery into a repeatable product. Golden paths combined infrastructure, security, deployment, and operational standards into a paved route that developers could use without learning every platform detail.

AI-native workloads do not invalidate that model. They expose where it stops too early.

A conventional golden path typically starts with source code and ends with a running service. An AI-native product depends on a wider chain: governed data, accelerator capacity, models and prompts, evaluation evidence, inference controls, agent identities, external tools, and continuous cost and risk feedback. If each of those capabilities arrives through a separate specialist portal, the organization has not created an AI platform. It has created another integration problem.

The next platform contract should therefore extend the golden path rather than build a parallel AI silo. The goal is not to hide every AI decision behind automation. It is to make safe defaults easy, exceptions explicit, and every promoted artifact traceable.

The application delivery contract is no longer enough

Platform Engineering 1.0 concentrated on a familiar delivery unit: an application packaged as a container, deployed through a pipeline, and operated with standard observability and security controls. That remains valuable, but AI changes both the workload and its consumers.

ML engineers need experiment tracking, model registries, feature and data access, and specialized compute. Application teams need stable inference endpoints and predictable latency. Security teams need controls for model provenance, prompt injection, data leakage, and non-human identities. FinOps teams need to attribute expensive training and inference usage. AI agents themselves become platform consumers that request tools, credentials, and runtime actions.

The CNCF discussion of evolving platform engineering for AI-native workloads captures this expansion through capabilities such as GPU and TPU allocation, model serving, MCP gateways, agentic guardrails, embedded FinOps, and policy-driven governance. The important organizational point is that these should not become an isolated platform owned by a small AI team. They should become extensions of the same product model, interfaces, and control philosophy used by the enterprise platform.

Define a platform contract, not a catalog of tools

A platform contract describes what a product team can request, what evidence it must provide, what the platform guarantees, and which controls are automatically applied. It is stronger than a service catalog entry and more flexible than a single mandatory implementation.

For an AI-native workload, that contract should cover at least six dimensions.

1. Governed data access

The path should make data classification, residency, retention, and permitted use visible before a workload reaches production. A request for a dataset should resolve to an approved identity, purpose, environment, and audit trail. The platform can automate access, but the product team remains accountable for whether the data is appropriate for the use case.

2. Compute and accelerator intent

Teams should request capabilities rather than hard-code a particular GPU model into every manifest. The contract can express workload class, memory, performance objective, duration, geographic constraints, and cost ceiling. Kubernetes mechanisms such as Dynamic Resource Allocation can support more structured resource claims, but the platform still needs policy for quotas, scarcity, preemption, and approved hardware profiles.

3. Model, prompt, and artifact provenance

Container images are not the only production artifacts. The platform must track model version, source, license, evaluation result, prompt bundle, retrieval configuration, tool definitions, and deployment policy. Promotion should be based on an immutable set of linked artifacts, not a model name copied into an environment variable.

4. Evaluation as a release gate

AI quality is probabilistic and context-dependent. A successful build does not prove production fitness. Golden paths should provide standard evaluation suites for task quality, safety, latency, robustness, and cost. Teams can add domain-specific tests, while the platform supplies the execution environment, evidence format, thresholds, and promotion workflow.

5. Runtime identity and guardrails

An inference service or autonomous agent needs a workload identity, scoped data access, approved tools, network boundaries, and observable policy decisions. The contract should distinguish a human user’s authority from an agent’s delegated authority. It should also define what happens when a model, tool, or policy is unavailable rather than allowing silent fallback to an uncontrolled path.

6. Cost and operational accountability

AI infrastructure introduces different cost behavior from ordinary stateless services. Training jobs can consume scarce capacity in bursts. Inference cost depends on model choice, token volume, batching, cache efficiency, and service-level objectives. Cost attribution and budgets should therefore be part of provisioning and release decisions, not a dashboard reviewed after the invoice arrives.

What an AI-native golden path looks like

A useful golden path follows the product lifecycle rather than exposing a collection of disconnected infrastructure forms.

  1. Declare the workload. The team selects an archetype such as batch training, online inference, retrieval-augmented generation, or tool-using agent. It declares data class, expected scale, latency objective, risk tier, and ownership.
  2. Provision an isolated workspace. The platform creates namespaces, identities, network boundaries, secrets references, storage, accelerator claims, quotas, and standard telemetry.
  3. Develop with approved building blocks. Teams consume versioned model endpoints, registries, feature services, MCP or tool gateways, and evaluation templates through stable APIs.
  4. Produce evidence. CI records model and data lineage, software dependencies, evaluation results, policy decisions, security findings, and predicted operating cost.
  5. Promote as a release set. GitOps promotes the linked application, model, prompt, policy, and tool configuration together. A rollback restores the complete known-good set.
  6. Operate with continuous feedback. Runtime telemetry covers service health, model quality indicators, policy denials, data drift, tool calls, accelerator utilization, and unit economics.

This lifecycle gives specialists room to innovate without forcing every product team to assemble the control plane themselves.

Avoid the separate AI platform trap

A dedicated AI enablement team may be necessary, but a separate delivery system should not be the default. Parallel identity models, pipelines, policy engines, and observability stacks increase cost and weaken governance. They also create a handoff between application engineers and AI specialists exactly where the product needs shared accountability.

A better operating model separates platform ownership by capability while preserving one product contract:

  • The core platform team owns common interfaces, workload identity, delivery workflows, policy integration, and the developer experience.
  • The AI platform capability team owns model-serving patterns, evaluation services, accelerator profiles, registries, and AI-specific runtime controls.
  • Data teams own governed data products and access semantics.
  • Security and risk teams define control objectives and approval boundaries as policy and evidence requirements.
  • Product teams own business fitness, domain evaluations, production outcomes, and accepted residual risk.

The teams collaborate through APIs, schemas, policy bundles, and service-level objectives rather than tickets and undocumented exceptions.

Measure whether the contract creates value

An AI-native platform should not be measured by the number of services in its catalog. Measure whether teams can deliver trustworthy outcomes faster.

  • Time from approved use case to first governed experiment
  • Time from candidate model to production release
  • Percentage of releases with complete model, data, prompt, and policy provenance
  • Evaluation failure escape rate
  • Percentage of agent tool calls using approved identities and gateways
  • Accelerator utilization and queue time by workload class
  • Inference cost per business transaction
  • Rollback time for a complete AI release set
  • Adoption and exception rates for each golden path

These metrics reveal whether the platform improves flow and control together. High adoption with slow delivery signals an overloaded path. Fast delivery with weak evidence signals unmanaged risk.

A practical 90-day starting point

Do not begin by designing a universal AI platform. Choose one real workload and use it to define the minimum viable contract.

Days 1–30: map the lifecycle

  • Select one representative AI product with a committed owner.
  • Map every artifact, identity, environment, approval, and operational dependency.
  • Classify which existing platform capabilities can be reused and where AI-specific gaps exist.
  • Define the workload’s risk tier, evaluation evidence, and cost objectives.

Days 31–60: build one vertical path

  • Create one workload template and governed workspace.
  • Connect model and prompt provenance to the existing GitOps release flow.
  • Add standard telemetry, policy checks, evaluation execution, and cost labels.
  • Document escape hatches with owners, expiry dates, and review requirements.

Days 61–90: prove and productize

  • Run a production-like release and rollback.
  • Measure lead time, evidence completeness, operational quality, and unit cost.
  • Interview the platform consumers and remove unnecessary steps.
  • Publish the contract as versioned schemas, APIs, examples, and service-level expectations.

The platform becomes the organizational control surface

AI-native platform engineering is not a race to add GPUs and model registries to an internal portal. It is the work of extending a proven product contract across a more complex value stream.

The strongest platforms will preserve what already works: product thinking, self-service, golden paths, policy automation, and composable cloud-native interfaces. They will add the missing contracts for data, models, evaluations, agents, specialized compute, and cost. That approach avoids a new silo while giving teams a credible path from experimentation to governed production.

Sources

Veröffentlicht am

OpenTelemetry Fleet Management: Why OpAMP Belongs in the Enterprise Observability Control Plane

OpenTelemetry can standardize how an enterprise collects and exports telemetry, but standardization alone does not make the collection layer operable.

At small scale, teams can manage Collector configuration through deployment manifests, virtual machine tooling, or a handful of automation scripts. At enterprise scale, the fleet becomes heterogeneous: Kubernetes DaemonSets, centralized gateways, virtual machines, laptops, point-of-sale devices, edge systems, and embedded environments. Different teams deploy the agents, while a central observability group remains accountable for data quality and service reliability.

That creates a control-plane problem. The organization needs to know which agents exist, what they are running, whether their configuration is current, whether a rollout succeeded, and how to recover without losing telemetry. The Open Agent Management Protocol, or OpAMP, provides a vendor-neutral protocol for that management relationship.

The strategic point is bigger than remote configuration. OpAMP belongs in the enterprise observability control plane because telemetry collection is production infrastructure. It needs identity, desired state, health feedback, controlled rollout, auditability, and rollback just like any other critical fleet.

Telemetry standardization exposes the management gap

OpenTelemetry adoption often begins with a sensible objective: remove proprietary instrumentation and normalize traces, metrics, and logs around open standards. The Collector becomes a flexible processing and export layer between workloads and one or more observability backends.

Success creates a new operating challenge. Collector configurations diverge by environment and team. Components run different versions. Credentials rotate at different times. Pipelines fail silently or begin dropping data. A change that looks safe in a development cluster can overload a regional gateway or remove a critical security log source.

GitOps helps with Kubernetes-managed Collectors, but it does not automatically cover agents on virtual machines, workstations, edge locations, or devices. It also tells the platform what was declared, not necessarily what every agent loaded or whether the resulting pipeline is healthy.

An enterprise control plane must connect declared intent with runtime evidence across the entire fleet.

What OpAMP actually provides

The OpenTelemetry specification describes OpAMP as a network protocol for remotely managing large fleets of data collection agents. It is vendor-agnostic and supports communication between an OpAMP server and clients associated with managed agents.

Core capabilities include:

  • Reporting agent identity, description, version, capabilities, and health
  • Receiving and acknowledging remote configuration
  • Reporting effective configuration and configuration status
  • Reporting package or component inventory
  • Receiving package update offers where the implementation supports them
  • Establishing bidirectional management communication over WebSocket or HTTP

OpAMP is a protocol, not a complete fleet-management product. It does not decide who may approve a production configuration, how rollout rings are selected, what policy is acceptable, or how a failed change should be escalated. Those are control-plane responsibilities that an enterprise platform must implement around the protocol.

The specification is currently marked beta. Newer Collector management work, including an alpha OpAMP Gateway Extension discussed by the CNCF, is promising but should be treated according to its maturity. Protocol adoption and production rollout should be deliberately separated from assumptions about experimental components.

The observability control plane needs a clear contract

A useful control plane maintains two views of every managed agent.

Desired state describes what the organization intends: approved Collector version, component set, configuration bundle, certificates, export destinations, and rollout assignment.

Observed state describes what the agent reports: identity, capabilities, effective configuration, health, errors, version, and last successful communication.

The difference between these views is configuration drift. Drift is not automatically a failure. An agent may be offline, a rollout may be paused, or a local emergency override may be permitted. The control plane should classify the difference, assign an owner, and decide whether to reconcile, roll back, or escalate.

This is why OpAMP complements rather than replaces GitOps. Git remains the reviewable source of approved configuration. OpAMP provides a standardized delivery and feedback channel for agents that cannot all be managed through the same deployment mechanism.

A reference enterprise architecture

A practical architecture separates policy, rollout orchestration, and protocol transport.

  1. Configuration repository. Versioned Collector templates, component allow lists, routing policy, environment overlays, and rollout metadata are reviewed through pull requests.
  2. Build and validation service. Every bundle is parsed, semantically validated, policy-checked, and tested against representative telemetry before promotion.
  3. Fleet inventory. The platform records agent identity, owner, environment, workload class, capabilities, current version, desired version, and health.
  4. Rollout controller. A change is assigned to cohorts, advanced through rings, paused on thresholds, and linked to an immutable configuration revision.
  5. OpAMP server. The server communicates desired state to clients and receives acknowledgements and status. It should not become the only system of record for policy decisions.
  6. Managed agents. Collectors or supervisors authenticate to the control plane, apply supported changes, and report effective state and health.
  7. Control-plane observability. The management system emits its own metrics, logs, and traces to an independent path so a fleet failure remains visible.

This architecture keeps configuration governance in familiar enterprise workflows while using OpAMP for standardized fleet interaction.

Identity is the first security boundary

A remote management channel can change what telemetry is collected, where it is sent, and which components execute. It is therefore a high-value security boundary.

Each client needs a stable identity tied to an owner and expected environment. Transport encryption is necessary but not sufficient. The server must authorize what that identity may receive, which cohort it belongs to, and whether it can accept sensitive configuration.

Recommended controls include:

  • Mutual authentication with short-lived, automatically rotated credentials
  • Per-agent or narrowly scoped workload identities rather than shared fleet secrets
  • Authorization by tenant, environment, geography, and workload class
  • Signed or integrity-protected configuration artifacts
  • Strict separation between configuration authors, approvers, and rollout operators
  • Audit records linking every server instruction to a reviewed revision and actor
  • Egress restrictions so agents communicate only with approved management and telemetry endpoints
  • Safe local behavior when the management server is unavailable

The server also needs protection from compromised agents. Rate limits, message-size limits, tenant isolation, replay resistance, input validation, and anomaly detection should be part of the threat model.

Configuration rollout should look like progressive delivery

Collector configuration can affect the visibility of an entire production estate. Treating a fleet-wide change as a simple push is an operational risk.

A safer workflow uses progressive rollout rings:

  1. Validation. Parse the configuration, resolve components, verify endpoints, run policy checks, and exercise representative telemetry.
  2. Development cohort. Apply the revision to disposable or low-risk agents and verify configuration acknowledgement.
  3. Canary cohort. Select a small production group that represents important environments and traffic patterns.
  4. Regional or workload rings. Expand only while health, drop rate, queue pressure, and backend load remain within thresholds.
  5. Fleet completion. Record coverage and identify offline or incompatible agents as explicit exceptions.
  6. Rollback. Restore the last known-good revision automatically when defined safety conditions fail.

A rollout is not successful because the server sent a configuration. It is successful when the intended agents report the expected effective state and the telemetry pipeline remains healthy.

Observe the observability fleet

The Collector layer is part of the monitoring system, so its management telemetry must not disappear into the same failure domain.

Track at least:

  • Active, offline, unknown, and quarantined agents
  • Desired-versus-effective configuration drift
  • Configuration acknowledgement and failure rates
  • Rollout duration and rollback frequency
  • Agent version and component-version distribution
  • Telemetry receive, drop, retry, queue, and export-failure rates
  • Credential age and failed authentication attempts
  • Management-channel latency and reconnect rate
  • Coverage by business service, environment, and data type

Business-level objectives matter as well. How quickly can the organization deploy a new security log source? How long does it take to revoke a compromised exporter credential? What percentage of critical services has a healthy, policy-compliant collection path?

An operating model for shared ownership

Fleet management spans organizational boundaries. Clear ownership prevents the control plane from becoming either an unresponsive central bottleneck or an uncontrolled self-service system.

  • The observability platform team owns the protocol service, supported agent profiles, configuration schemas, rollout automation, and service-level objectives.
  • Security owns control objectives, management-plane threat modeling, credential requirements, and sensitive destination policy.
  • Service and infrastructure teams own agent coverage, local dependencies, and declared business criticality.
  • Backend owners publish capacity constraints and compatibility requirements.
  • A change advisory model is encoded through risk tiers, automated evidence, and approval rules rather than a universal manual meeting.

Teams should be able to request supported pipelines and processors through a controlled interface. They should not need permission for every low-risk change, but they also should not be able to redirect enterprise telemetry to an arbitrary endpoint.

A staged adoption plan

Phase 1: establish inventory and evidence

  • Enumerate Collector deployments and other managed agents.
  • Assign ownership, environment, and criticality.
  • Define a small set of supported configurations and component versions.
  • Measure current drift, rollout time, and blind spots before adding remote control.

Phase 2: introduce OpAMP in read-oriented mode

  • Connect a non-critical cohort.
  • Collect agent descriptions, versions, effective configuration, and health.
  • Validate identity and tenant boundaries.
  • Compare observed state with the Git-approved desired state.

Phase 3: controlled configuration delivery

  • Enable remote configuration for one standardized agent profile.
  • Use signed revisions, canary rings, automated thresholds, and rollback.
  • Exercise server outage, invalid configuration, expired credentials, and incompatible-agent scenarios.

Phase 4: expand deliberately

  • Add heterogeneous environments and additional agent capabilities.
  • Integrate package updates only after configuration delivery is reliable.
  • Publish service objectives and an exception process.
  • Keep experimental extensions behind explicit maturity and risk gates.

Standard telemetry needs standard operations

OpenTelemetry solves an important portability problem, but enterprises also need a portable way to operate the collection fleet. OpAMP creates the protocol foundation for that control plane.

The durable design is not a central server that can push arbitrary files. It is a governed system that connects reviewed intent to agent identity, progressive rollout, effective state, health evidence, and safe recovery. Organizations that build those capabilities can scale OpenTelemetry without replacing proprietary telemetry agents with a new collection layer that is open but operationally opaque.

Sources

Veröffentlicht am

GitOps for AI Agents: Why Prompts, Tools, and Policies Belong in Your Platform Repository

AI agents are increasingly moving from experiments into production workflows. They can inspect systems, call tools, change infrastructure, open pull requests, and trigger operational actions. Yet many teams still manage the most important parts of an agent—its system prompt, tool permissions, output contract, and safety rules—as scattered text in notebooks, environment variables, or application code.

That is not just inconvenient. It is a governance problem.

If agent configuration influences production behavior, it should be managed like any other form of production configuration: declarative, versioned, reviewed, testable, and reversible. This is where GitOps becomes relevant—not as another fashionable label, but as a practical operating model for agentic systems.

Agent configuration is production behavior

For a conventional service, teams already treat deployment manifests, network policies, resource limits, and feature flags as controlled artifacts. An AI agent adds another behavioral layer:

  • the system prompt defines role, boundaries, and decision priorities;
  • the tool list determines which actions the agent can perform;
  • the output schema defines what downstream systems may trust;
  • policy bundles decide which actions are allowed, denied, or escalated;
  • model and routing settings affect cost, latency, and risk;
  • confidence and blast-radius thresholds determine when a human must intervene.

A change to any of these elements can alter production outcomes without changing a single line of traditional application code. Treating them as informal configuration creates an audit gap: teams may know which container image ran, but not which instructions or tool permissions shaped the agent’s decision.

What GitOps adds

The OpenGitOps principles describe desired state as declarative, versioned and immutable, automatically pulled, and continuously reconciled. Applied to agents, these principles create a clear chain from intent to runtime behavior.

A practical model looks like this:

  1. Agent configuration is stored in Git as structured data.
  2. A pull request shows the exact behavioral change.
  3. Automated checks validate schemas, policies, permissions, and evaluation results.
  4. Reviewers approve the change based on ownership and risk.
  5. A GitOps controller reconciles the approved state into the runtime platform.
  6. Telemetry confirms which version is active and how it behaves.
  7. A rollback restores the last known-good configuration when required.

This is already being applied in real cloud-native agent platforms. In a CNCF case study from Orange Innovation, each agent’s system prompt, tool list, and output schema is represented as a Kubernetes Custom Resource and reconciled from Git through Argo CD. Safety policies live in the same repository, making promotion code-reviewed, auditable, and reversible.

What should live in Git?

The goal is not to put every piece of runtime context into a repository. Git should contain the stable desired state that governs the agent.

Good candidates

  • system prompts and instruction templates;
  • allowed and denied tool definitions;
  • input and output schemas;
  • policy-as-code bundles;
  • model selection and fallback rules;
  • human-approval thresholds;
  • resource limits and deployment settings;
  • evaluation datasets and acceptance thresholds;
  • ownership metadata and escalation routes.

What should not live in Git?

  • API keys, tokens, and credentials;
  • personal or customer-sensitive conversation data;
  • short-lived runtime context;
  • unfiltered model traces containing confidential data;
  • mutable operational state that belongs in a database or event stream.

Secrets should be referenced through a secret-management system. Dynamic context should be retrieved through controlled tools with explicit identity, authorization, and audit trails.

An illustrative Kubernetes resource

Kubernetes Custom Resources provide one possible way to model agent desired state. The following example is illustrative rather than a proposed standard:

apiVersion: agents.platform.it-stud.io/v1alpha1
kind: AgentConfiguration
metadata:
  name: incident-reviewer
spec:
  promptRef: prompts/incident-reviewer-v12
  modelPolicy:
    primary: approved-enterprise-model
    fallback: approved-low-latency-model
  tools:
    allow:
      - read-observability-data
      - create-incident-ticket
    deny:
      - execute-production-change
  outputSchemaRef: schemas/incident-review-v3.json
  policyBundleRef: policies/soc-reviewer-v8
  humanApproval:
    requiredFor:
      - customer-facing-assets
      - identity-systems
      - actions-above-blast-radius-threshold

The value is not the YAML itself. The value is that the desired behavior becomes visible, reviewable, and reconcilable. A platform controller can translate this resource into runtime configuration while policy engines validate what teams are allowed to change.

The pull request becomes a governance control

A prompt review should not be treated like a copy-editing exercise. It is closer to reviewing infrastructure or authorization policy.

Different changes need different reviewers:

  • domain owners review whether instructions reflect the intended business process;
  • platform teams review runtime, deployment, and operational impact;
  • security teams review tool permissions, policy rules, identity, and blast radius;
  • AI engineers review model behavior, schemas, and evaluation results.

Branch protection and CODEOWNERS can turn this responsibility model into an enforceable workflow. A tool-permission change may require security approval, while a wording clarification within an existing boundary may only require the domain owner.

CI must test behavior, not just syntax

Schema validation is necessary but insufficient. An agent configuration can be valid YAML and still create unsafe or ineffective behavior.

A useful CI pipeline should combine:

  • schema and policy validation;
  • checks for forbidden tools or excessive permissions;
  • prompt-injection and adversarial test cases;
  • regression evaluations against representative scenarios;
  • cost and latency budgets;
  • output-schema conformance;
  • evidence that required human escalation still occurs.

The result should be an evaluation report attached to the pull request. Reviewers then see not only what changed, but how the agent’s measured behavior changed.

Deployment needs progressive delivery

GitOps makes rollback possible, but production agent changes should still be introduced gradually. A prompt or policy update can pass offline evaluations and fail under real operational conditions.

Platform teams can apply familiar delivery patterns:

  • shadow mode, where the new version makes decisions without executing them;
  • canary rollout to a limited workload or user group;
  • automatic rollback on quality, safety, latency, or cost regression;
  • version labels in traces so behavior can be tied to the exact Git revision;
  • human approval for changes that expand tool access or blast radius.

This is where agent operations begin to look less like prompt experimentation and more like mature platform engineering.

A practical operating model

Teams do not need a new organizational silo for every agent. They need clear contracts between existing responsibilities.

  • Domain teams own desired outcomes and business constraints.
  • AI engineering owns agent contracts, evaluations, and model behavior.
  • Platform engineering owns the runtime, GitOps reconciliation, observability, and deployment controls.
  • Security and risk own policy requirements, privileged actions, and evidence.

Machine-readable contracts—schemas, policies, Custom Resources, and evaluation thresholds—reduce coordination overhead. Teams can evolve their area without relying on undocumented meetings or hidden configuration.

A 30-day starting plan

  1. Inventory: identify production agents and locate their prompts, tools, policies, and schemas.
  2. Structure: move stable behavioral configuration into a versioned repository without migrating secrets or sensitive runtime data.
  3. Protect: add CODEOWNERS, branch protection, and approval requirements for high-risk fields.
  4. Validate: introduce schema checks, policy tests, and a small regression evaluation suite.
  5. Reconcile: automate deployment through an existing GitOps controller or equivalent reconciliation process.
  6. Observe: attach configuration version, model version, tool calls, cost, latency, and escalation outcomes to telemetry.
  7. Roll back: test restoration of the last known-good configuration before the first production incident.

Conclusion

AI agents should not be governed through scattered prompts and tribal knowledge. The configuration that shapes their behavior belongs in the same disciplined operating model used for other production systems.

GitOps provides a practical foundation: declared intent, version history, peer review, automated validation, continuous reconciliation, and fast rollback. Combined with policy-as-code, behavioral evaluations, progressive delivery, and human approval boundaries, it gives platform teams a credible way to scale agentic systems without losing control.

The core principle is simple: if a configuration change can alter what an agent is allowed to decide or do, it deserves the same engineering rigor as a production code change.

Sources and further reading

Veröffentlicht am

llm-d: Kubernetes-Native Distributed LLM Inference and CNCF’s Answer to the AI Inference Gap

For the last few years, Kubernetes has been the default control plane for almost everything except one increasingly important workload: large language model inference. Teams that adopted Kubernetes for microservices, batch jobs, and stateful systems often ran their GenAI inference on bespoke setups, vendor platforms, or hand-tuned vLLM deployments glued together with custom scripts. That gap is now closing.

At KubeCon EU 2026, the CNCF accepted llm-d into its Sandbox. Backed by Red Hat, Google, IBM, CoreWeave, and NVIDIA, llm-d is a purpose-built, Kubernetes-native framework for distributed LLM inference. It is not another model server competing with vLLM. It is the orchestration and serving layer that turns a fleet of accelerators into a coherent, scalable inference platform on Kubernetes.

This matters because the economics and operational reality of LLM inference are fundamentally different from traditional web services. A single large model can exceed the memory of one GPU. Request patterns are bursty. The compute profile of generating the first token is completely different from generating the rest. Treating inference like a stateless HTTP service leaves enormous performance and cost on the table. llm-d is the CNCF community’s answer to that problem.

The Inference Gap: Why Kubernetes Needed a Native Answer

According to the CNCF’s 2026 survey, around 66% of organizations now run GenAI inference on Kubernetes. That adoption happened faster than the tooling matured. Most teams stitched together their own stack: a model server such as vLLM, a custom Deployment or StatefulSet, an Ingress or Gateway in front, a homegrown autoscaler keyed off queue depth, and a lot of YAML.

This works until it doesn’t. The problems show up at scale:

  • Model size vs. accelerator memory: Large models must be sharded across multiple GPUs or nodes, which standard Deployments do not coordinate well.
  • Prefill vs. decode imbalance: The compute-heavy prefill phase and the memory-bandwidth-bound decode phase compete for the same resources when colocated.
  • Scheduling fragility: Multi-GPU inference pods need all their resources at once. Partial scheduling wastes expensive accelerators.
  • Routing blindness: Standard HTTP load balancing does not understand KV-cache locality, model affinity, or queue depth.
  • Cost opacity: Without per-accelerator metrics, FinOps for inference is guesswork.

llm-d addresses these as first-class concerns rather than afterthoughts. It aligns with the emerging Kubernetes AI Requirements (KARs) and positions itself as the reference inference stack for the cloud-native ecosystem.

Prefill/Decode Disaggregation: The Core Idea

The single most important architectural concept in llm-d is the disaggregation of the prefill and decode phases of inference.

When an LLM processes a request, it first reads the entire prompt and builds a key-value (KV) cache. This is the prefill phase, and it is compute-bound: it benefits from raw GPU throughput. Then the model generates tokens one at a time, each step depending on the KV cache. This is the decode phase, and it is largely memory-bandwidth-bound and latency-sensitive.

When both phases run on the same GPU, they interfere. A long prompt being prefilled can stall token generation for other requests, hurting time-to-first-token and inter-token latency simultaneously. Disaggregation separates these phases onto different pools of accelerators that can be scaled and tuned independently.

Phase Characteristic Bottleneck Scaling Strategy
Prefill Processes the full prompt, builds KV cache Compute (FLOPs) Scale for throughput on high-FLOP accelerators
Decode Generates tokens iteratively Memory bandwidth, latency Scale for concurrency and low latency

By splitting these across nodes, llm-d lets platform teams right-size each pool. You can throw high-throughput accelerators at prefill and optimize a separate pool for low-latency decode, transferring the KV cache between them. The result is better GPU utilization and more predictable latency under load.

How llm-d Fits the Kubernetes Stack

llm-d is not a monolith. It is designed to compose with the newest Kubernetes primitives for AI workloads. This is what makes it cloud-native rather than just another inference server wrapped in a container.

Dynamic Resource Allocation (DRA)

Kubernetes v1.36 matured Dynamic Resource Allocation, which replaces the aging device plugin framework for GPUs and accelerators. NVIDIA and Google have contributed CNCF-donated DRA drivers that act as the accelerator abstraction layer. llm-d uses DRA to request GPUs declaratively, with structured parameters the scheduler and autoscaler can actually understand. This means topology-aware allocation, partitionable devices, and cleaner multi-accelerator scheduling.

Gang Scheduling

Distributed inference needs all its pods at once. A model sharded across four GPUs is useless with three. llm-d relies on gang scheduling, available in the Kubernetes v1.36 workload-aware scheduling features, to ensure that a distributed inference deployment either gets all its resources or waits, rather than partially allocating and stranding expensive accelerators.

Kueue for Job Queuing

For multi-tenant inference pools and batch-style inference, llm-d integrates with Kueue. This brings quota management, fair sharing, and queuing across teams. Platform teams can define ClusterQueues that cap GPU budgets per team while keeping the pool efficiently shared.

Gateway API Inference Extension

Above llm-d sits the Gateway API Inference Extension (GIE), which provides intelligent routing for LLM traffic. Unlike standard HTTP routing, GIE understands inference-specific signals: model affinity, KV-cache locality, queue depth, and load. It routes requests to the right pool and the right replica, which is essential once prefill and decode are disaggregated.

llm-d vs. vLLM, KServe, and Ray Serve

A common point of confusion is where llm-d sits relative to existing tools. It does not replace all of them; it orchestrates and complements them.

Tool Primary Role When to Use
vLLM High-performance model server / inference engine As the underlying engine, often used by llm-d itself
KServe General model serving framework on Kubernetes Mixed model types, classic ML plus LLMs, standardized CRDs
Ray Serve Python-native distributed serving Teams already invested in the Ray ecosystem
llm-d Kubernetes-native distributed LLM inference orchestration Large-scale LLM inference with prefill/decode disaggregation and multi-node scaling

The practical mental model: vLLM is the engine, llm-d is the distributed serving and orchestration layer that runs engines like vLLM across a fleet, and the Gateway API Inference Extension is the smart front door. KServe and Ray Serve remain valid choices, especially for mixed workloads, but llm-d is purpose-built for the specific challenges of large-scale LLM inference on Kubernetes.

Observability and Cost: Making Inference Accountable

One of the most underrated aspects of running inference at scale is knowing what it costs and how it performs per accelerator. llm-d treats observability as a requirement, aligning with AI Conformance expectations.

Key signals platform teams should capture:

  • Time-to-first-token (TTFT): Dominated by prefill and queueing; the primary latency SLO for interactive use.
  • Inter-token latency: Reflects decode performance and concurrency pressure.
  • Per-accelerator utilization: GPU compute and memory usage, exported via Prometheus.
  • Queue depth and batch size: Drive autoscaling decisions far better than CPU metrics.
  • Throughput (tokens/sec): The real unit of inference work.

With OpenTelemetry and Prometheus wired in, these metrics feed both autoscaling and FinOps. The cost question llm-d helps answer is concrete: is it cheaper to run distributed inference across several smaller accelerators with disaggregation, or to consolidate on fewer large GPU nodes? The answer depends on model size, request mix, and latency targets, but for the first time the data to decide is available natively.

Security and Multi-Tenancy

Inference pools are expensive shared resources, which makes isolation important. llm-d supports multi-tenant inference pools with workload isolation and RBAC. Combined with Kueue quotas and namespace-scoped policies, platform teams can let multiple AI teams share a GPU fleet without one team starving another or accessing another team’s models. This is the same multi-tenancy discipline platform teams already apply to compute, extended to accelerators.

A Production Readiness Checklist

llm-d is in the CNCF Sandbox, which means it is early. Sandbox status signals direction and community backing, not production maturity. Platform teams evaluating it should be deliberate.

  1. Validate your Kubernetes version. You need v1.36-era features: mature DRA, gang scheduling, and workload-aware scheduling. Confirm your managed provider (EKS/GKE/AKS) exposes the required feature gates and DRA drivers.
  2. Confirm accelerator drivers. Ensure the NVIDIA or Google DRA drivers are installed and supported on your node pools.
  3. Start with a single model and pool. Prove the basic serving path before introducing prefill/decode disaggregation across nodes.
  4. Wire observability first. Export TTFT, inter-token latency, throughput, and per-GPU metrics to Prometheus before scaling up.
  5. Introduce the Gateway API Inference Extension early. Smart routing is what makes disaggregation pay off.
  6. Layer in Kueue for multi-tenancy. Define GPU quotas per team before opening the pool to multiple consumers.
  7. Run a cost comparison. Benchmark distributed inference against your current setup with real traffic, not synthetic load.
  8. Plan for sandbox churn. APIs may change. Pin versions, track releases, and avoid hard-coupling your platform to unstable interfaces.

Why This Is a Platform Engineering Story

It is tempting to file llm-d under „AI infrastructure“ and leave it to ML teams. That would be a mistake. The whole point of llm-d is to make distributed LLM inference a self-service IDP primitive. Instead of every AI team building bespoke serving stacks, the platform team offers inference as a paved road: declare a model, a pool, and an SLO, and the platform handles GPU allocation, scheduling, routing, scaling, and observability.

This is the same shift platform engineering brought to application deployment, now applied to inference. The golden path for an AI team becomes a governed, observable, cost-aware inference service rather than a pile of custom YAML and tribal knowledge.

The Bottom Line

llm-d represents the cloud-native ecosystem catching up to where AI workloads actually are. By disaggregating prefill and decode, integrating with DRA, gang scheduling, Kueue, and the Gateway API Inference Extension, and treating observability and multi-tenancy as requirements, it offers a coherent answer to the inference gap on Kubernetes.

It is early, and Sandbox status means platform teams should pilot rather than bet the farm. But the backing from Red Hat, Google, IBM, CoreWeave, and NVIDIA, plus alignment with the Kubernetes AI Requirements, makes llm-d the most credible candidate to become the standard distributed inference layer for Kubernetes. For platform teams whose organizations are already running GenAI on Kubernetes, now is the time to start evaluating it, building the observability foundation, and planning the golden path for inference-as-a-service.

Veröffentlicht am

Internal Developer Portals 2.0: How AI Copilots Inside Backstage and Port Are Transforming Developer Self-Service

Internal Developer Portals have spent the past three years earning their place in the platform engineering stack. Backstage — now a CNCF Graduated project — established the blueprint: a service catalog, software templates, TechDocs, and a plugin ecosystem exceeding 900 integrations. For many organizations, that was enough. But static catalogs have hit a ceiling. Developers still context-switch between Backstage, Slack, their IDE, and a dozen dashboards to scaffold a service, request infrastructure, or troubleshoot an incident. The portal that was supposed to unify developer experience became just another tab.

2025 and 2026 have introduced a different paradigm: AI copilots embedded directly inside IDPs. Not chatbots bolted onto the side, but intelligent agents that understand your service catalog, your golden paths, and your organizational policies — and let developers interact with infrastructure through natural language instead of form-driven UIs. This is Internal Developer Portals 2.0, and it changes the economics of platform engineering.

From Catalog-Centric to Action-Centric Portals

The first generation of IDPs was catalog-centric. You browsed a list of services, looked up ownership, maybe triggered a pre-built template. The developer experience was better than nothing, but it still required knowing where to click and which template to use. For a senior engineer who helped build the portal, that was fine. For a new hire on day three, it was another maze.

Action-centric IDPs flip the model. Instead of navigating a catalog hierarchy, a developer types:

"Deploy my payment-service to staging with the new database migration"

The AI copilot inside the portal understands the intent, resolves the service from the catalog, identifies the correct deployment pipeline, checks RBAC policies, and either executes or presents a confirmation step. The catalog is still there — it’s the knowledge backbone — but the interaction layer has fundamentally changed.

This isn’t speculative. Port has shipped an AI assistant that queries its internal software catalog and executes self-service actions through natural language. Cortex integrates LLM-driven recommendations directly into its scorecards. Humanitec has taken an API-first approach that makes AI orchestration a first-class integration pattern. Even Backstage itself is seeing community plugins that expose catalog data to AI agents via standardized protocols.

The Knowledge Graph Advantage

What makes IDP-embedded AI fundamentally different from a generic ChatGPT wrapper is context. An Internal Developer Portal already holds a rich knowledge graph:

  • Service dependencies: which services call which, what databases they use, what message queues connect them
  • Team ownership: who owns what, who’s on-call, escalation paths
  • Runbooks and documentation: operational playbooks indexed per service
  • Deployment history: what was deployed when, by whom, with what configuration
  • Scorecards: production readiness, security posture, cost allocation

When an AI copilot has access to this graph, its responses move from generic to surgical. Ask it "Why is checkout-service latency spiking?" and it can correlate recent deployments, check the dependency graph for upstream changes, pull relevant runbooks, and suggest specific remediation steps — all without the developer leaving the portal.

Compare this to ChatOps bots in Slack that operate with minimal context, or IDE-integrated copilots that understand your code but not your infrastructure. The IDP sits at the intersection of code, infrastructure, and organizational knowledge. That’s where AI adds the most leverage.

MCP Servers: The Bridge Between AI Agents and IDP APIs

The technical glue making this possible is increasingly the Model Context Protocol (MCP). Originally open-sourced by Anthropic in 2024 and now seeing broad adoption, MCP provides a standardized interface for AI agents to discover and invoke tools — including IDP APIs.

An MCP server wrapping your Backstage or Port API exposes capabilities like:

  • Querying the service catalog (list services, get owners, check dependencies)
  • Triggering software templates (scaffold a new microservice, provision a database)
  • Reading and updating scorecards
  • Executing self-service actions (deploy, rollback, scale)
  • Fetching TechDocs and runbooks for context

This decouples the AI layer from the IDP implementation. Your platform team maintains the MCP server as a thin adapter. The AI agent — whether it’s embedded in the portal UI, accessible via Slack, or running inside an IDE — connects through the same protocol. You get a single source of truth for what actions are available and what permissions govern them.

For teams already running Backstage, this is particularly powerful. The existing plugin ecosystem handles data aggregation; an MCP server adds an AI-native interaction layer on top without replacing the portal itself.

Scorecards Meet AI Analysis

Scorecards have been one of the quiet successes of the IDP movement. Tools like Backstage (via the Scorecards plugin), Port, and Cortex let platform teams define maturity criteria — production readiness, security compliance, documentation coverage, cost efficiency — and track every service against them.

AI transforms scorecards from passive dashboards into active recommendation engines:

  • Service maturity gaps: „Your order-service scores 62% on production readiness. Adding health check endpoints and configuring pod disruption budgets would bring it to 85%.“
  • Security posture: „Three services in the payments domain are running container images older than 90 days. Here are the specific CVEs affecting them.“
  • Cost optimization: „Based on CPU utilization patterns over the last 30 days, analytics-worker is over-provisioned by 3x. Recommended resource requests: 200m CPU, 256Mi memory.“

The shift is from „here’s your score“ to „here’s what to do about it.“ When combined with self-service actions, the AI can even generate the pull request to implement the recommendation — turning insight into action in a single interaction.

Dynamic Golden Paths: AI-Generated Templates

Golden paths — the blessed, paved roads for common developer tasks — have traditionally been static. Your platform team creates a service template, a database provisioning workflow, a CI/CD pipeline configuration. Developers pick from the menu.

AI-powered IDPs make golden paths dynamic. Instead of maintaining 15 slightly different service templates for different tech stacks and deployment targets, you maintain a smaller set of composable building blocks. The AI assembles them based on the developer’s intent:

"I need a new Go microservice with a PostgreSQL database, deployed to our EU region, with PII data handling compliance"

The copilot generates a tailored template that includes the correct Helm values for the EU cluster, enables encryption-at-rest annotations for PII compliance, configures the appropriate network policies, and sets up the CI/CD pipeline with the required security scanning stages. The golden path isn’t a fixed road anymore — it’s a GPS that calculates the route based on where you’re going.

This has real implications for template maintenance. Platform teams spend significant effort keeping templates current across Kubernetes versions, policy changes, and infrastructure updates. AI-generated templates that compose from maintained primitives reduce that burden substantially.

Incident Response: The Killer Use Case

If there’s a single scenario where IDP-embedded AI proves its ROI overnight, it’s incident response. Consider the typical flow today:

  1. Alert fires in PagerDuty or Opsgenie
  2. On-call engineer opens the monitoring dashboard
  3. Checks recent deployments in the CI/CD tool
  4. Looks up service ownership in the IDP
  5. Searches for relevant runbooks
  6. Correlates with dependency graph to identify blast radius
  7. Begins remediation

Steps 2 through 6 are pure context gathering — and they happen under pressure at 3 AM. An AI agent inside the IDP can perform all of them in seconds:

  • Correlate the alert with the service catalog entry
  • Identify recent changes (deployments, config updates, dependency upgrades)
  • Pull the relevant runbook and highlight the most likely remediation steps
  • Map the blast radius through the dependency graph
  • Suggest or auto-execute a rollback if the confidence is high enough

The on-call engineer still makes the decision, but the mean time to context drops from 15 minutes to 15 seconds. For organizations running hundreds of microservices, that’s not a nice-to-have — it’s a competitive advantage.

Developer Experience Metrics: Measuring What Matters

AI-powered IDPs also change how we measure developer experience. The DORA metrics (deployment frequency, lead time for changes, change failure rate, mean time to recovery) and the SPACE framework (satisfaction, performance, activity, communication, efficiency) are becoming first-class citizens in IDP dashboards.

The AI layer adds predictive and diagnostic capabilities:

  • Trend analysis: „Deployment frequency for the checkout team has dropped 30% over the past sprint. The primary bottleneck appears to be flaky integration tests in the payment-gateway pipeline.“
  • Correlation: „Teams using the v3 service template have 40% lower change failure rates than those on v2. Consider migrating remaining v2 services.“
  • Forecasting: „Based on current velocity, the platform migration will complete in Q3 — two weeks later than planned. The blocker is database schema migrations for three legacy services.“

This is where platform engineering ROI becomes measurable. When you can demonstrate that AI-assisted self-service reduces time-to-production for new services from five days to four hours, the investment case writes itself.

Backstage’s Plugin Ecosystem vs. AI-Native Platforms

The market is splitting into two camps, and platform teams need to understand the tradeoffs:

Dimension Backstage + AI Plugins AI-Native Platforms (Port, Cortex)
Flexibility 900+ plugins, infinite customization Fewer but deeper integrations
AI integration Community-driven, via MCP/plugins Built-in, first-class AI features
Maintenance burden High (self-hosted, plugin compatibility) Lower (SaaS, managed updates)
Data ownership Full control (self-hosted) Vendor-dependent
Time to value Weeks to months Days to weeks
Vendor lock-in Low (CNCF, open source) Moderate to high
Knowledge graph depth As deep as you build it Pre-built entity models

Neither approach is universally better. If your organization has strong platform engineering capacity and wants full control, Backstage with AI plugins and MCP servers gives you maximum flexibility. If you want faster time-to-value and your team is lean, an AI-native platform like Port gets you to production-grade IDP faster — at the cost of some flexibility and data sovereignty.

ChatOps vs. IDP-Embedded AI vs. IDE Copilots

It’s worth clarifying where IDP-embedded AI fits relative to other AI integration points:

  • ChatOps (Slack/Teams bots): Good for notifications and simple commands. Limited context about your infrastructure. Works well for quick queries but struggles with complex multi-step workflows.
  • IDE-integrated copilots (GitHub Copilot, Cursor): Excellent for code generation. No awareness of your deployment topology, service catalog, or organizational policies. Wrong tool for infrastructure tasks.
  • IDP-embedded AI: Sits at the intersection of organizational knowledge, infrastructure state, and developer workflows. Best for self-service actions, incident response, and cross-cutting concerns that span multiple services.

The ideal setup uses all three — but the IDP is the orchestration layer. Your Slack bot calls the IDP’s AI capabilities through MCP. Your IDE copilot references the service catalog for context. The IDP is the brain; everything else is an interface.

Multi-Tenancy, RBAC, and Governance

Here’s where many early AI-in-IDP implementations fall short: governance. When an AI agent can trigger deployments, modify infrastructure, or scaffold services, you need the same (or stricter) access controls as your existing self-service workflows.

Critical requirements:

  • RBAC for AI actions: The AI copilot should inherit the requesting user’s permissions, not operate with elevated privileges
  • Audit trails: Every AI-initiated action must be logged with the full context — who asked, what was requested, what was executed, what was the outcome
  • Approval gates: Destructive or high-risk actions (production deployments, database migrations, security policy changes) should require human approval, even when AI-initiated
  • Multi-tenancy: In organizations with multiple teams sharing an IDP, AI actions must respect tenant boundaries. Team A’s copilot cannot access Team B’s secrets or deploy to Team B’s namespaces
  • Rate limiting: Prevent AI agents from executing runaway loops of infrastructure changes

Without these controls, you’re trading developer friction for security risk — not a trade worth making.

The Risks: What Can Go Wrong

Let’s be direct about the failure modes:

  • Hallucinated infrastructure configurations: An AI that generates a Kubernetes manifest with incorrect resource limits, missing security contexts, or wrong network policies can cause outages. Every AI-generated configuration must pass through the same validation pipelines (OPA/Kyverno, CI checks) as human-authored configs.
  • Insufficient audit trails: If an AI agent makes a change and the audit log only shows „AI modified resource X,“ you’ve lost forensic capability. Log the full chain: user prompt → AI interpretation → action taken → result.
  • Shadow IT acceleration: If self-service becomes too easy, developers spin up resources without proper tagging, cost allocation, or lifecycle management. AI-powered IDPs need to enforce organizational policies at the point of creation, not after the fact.
  • Over-reliance on AI recommendations: Scorecard suggestions and incident response playbooks should augment human judgment, not replace it. Build a culture where AI recommendations are validated, not blindly accepted.

Getting Started: A Practical Roadmap

If you’re running Backstage today and want to add AI capabilities, here’s a pragmatic path:

  1. Start with read-only: Build an MCP server that exposes your service catalog, scorecards, and documentation to an AI agent. Let developers query the catalog through natural language. Zero risk, immediate value.
  2. Add scorecard analysis: Connect the AI to your scorecard data and let it generate improvement recommendations. Still read-only, but now actively useful.
  3. Enable template generation: Allow the AI to compose software templates based on developer intent. Route the output through your existing PR review process.
  4. Introduce action execution: Wire up deployment, scaling, and provisioning actions with approval gates. Start with non-production environments.
  5. Extend to incident response: Connect alerting systems and let the AI perform context gathering and remediation suggestions during incidents.

Each step builds on the previous one, and each can be rolled back independently. The key is maintaining human oversight throughout — AI copilots augment your platform team, they don’t replace it.

The Bottom Line

Internal Developer Portals 2.0 aren’t about replacing Backstage or rebuilding your IDP from scratch. They’re about adding an intelligence layer that transforms the portal from a passive catalog into an active assistant. The service catalog becomes a knowledge graph. Templates become dynamic. Scorecards become recommendation engines. Incident response becomes proactive.

The organizations that get this right will see measurable improvements in developer productivity, onboarding speed, and operational resilience. The ones that don’t will keep maintaining static portals that developers tolerate rather than love.

The technology is ready. The protocols (MCP) are standardizing. The question isn’t whether AI belongs in your IDP — it’s how quickly you can integrate it without compromising the governance that makes your platform trustworthy.

Veröffentlicht am

The .de DNSSEC Meltdown: What Platform Teams Can Learn from Germany’s TLD Outage

TL;DR — On May 5 2026, DENIC pushed broken DNSSEC signatures into the .de zone. Because DNSSEC validation is a strict chain-of-trust model, every validating resolver on the planet began returning SERVFAIL for all .de domains. Millions of websites, APIs, and mail servers went dark. Resolvers that had deployed Serve-Stale (RFC 8767) and Negative Trust Anchors (RFC 7646) recovered within minutes; everyone else waited hours. This article breaks down the incident, the mitigation patterns, and the concrete steps platform teams should take so a single TLD mistake doesn’t take down their stack.

What Happened on May 5, 2026

At approximately 10:42 UTC on Monday, May 5, monitoring dashboards across Europe lit up. DNS resolution for .de domains — one of the world’s largest country-code TLDs, consistently ranking in the Top 5 at Cloudflare Radar — started failing en masse. The root cause: DENIC, the registry operator for .de, had published DNSSEC signatures that did not match the zone’s active Zone Signing Key (ZSK).

The timing was no coincidence. The faulty signatures surfaced during a scheduled ZSK rotation — one of the most operationally sensitive windows in DNSSEC key management. A misconfiguration in the signing pipeline meant that the new signatures were generated with a key that validating resolvers could not verify against the published DS records in the root zone. The result was catastrophic: the entire .de chain of trust was broken.

Within minutes, every DNSSEC-validating resolver worldwide — including Cloudflare’s 1.1.1.1, Google’s 8.8.8.8, and Quad9’s 9.9.9.9 — began returning SERVFAIL for queries to .de domains. Non-validating resolvers continued to work, which created a confusing split-brain situation where some users could reach German websites and others couldn’t, depending on their configured resolver.

The DNSSEC Chain of Trust: One Link Breaks, Everything Falls

To understand why a single registry mistake can have such a massive blast radius, you need to understand how DNSSEC validation works.

DNSSEC adds cryptographic signatures to DNS records. Resolvers verify these signatures by walking a chain of trust from the root zone (.) down through the TLD (.de) to the individual domain (example.de). Each level delegates trust to the next via DS (Delegation Signer) records. If any link in this chain produces an invalid signature, a validating resolver must return SERVFAIL. That’s not a bug — it’s the design. DNSSEC was built to prevent cache poisoning, and treating unverifiable answers as failures is the entire point.

The double-edged nature of this design becomes painfully clear during operator errors at the TLD level. When DENIC’s signatures broke, it wasn’t just one domain that failed — it was every single .de domain, regardless of whether the individual domain owner had done everything right. The TLD is a single point of cryptographic failure for all domains beneath it.

ZSK/KSK Rotation: The Critical Window

DNSSEC uses two types of keys: the Key Signing Key (KSK), which signs the DNSKEY RRset, and the Zone Signing Key (ZSK), which signs the actual zone data. ZSK rotations happen more frequently and involve a carefully choreographed dance: pre-publish the new key, wait for caches to expire, sign with the new key, remove the old one. Get any step wrong — wrong timing, wrong key reference, stale DS record — and you shatter the chain of trust. This is exactly what happened with .de.

How Major Resolvers Responded

The incident provided a real-world stress test for two mitigation techniques that the DNS community has been advocating for years: Serve-Stale and Negative Trust Anchors.

Serve-Stale (RFC 8767)

Serve-Stale allows a resolver to return expired (stale) cached records instead of failing with SERVFAIL when it cannot fetch a fresh, valid answer from upstream. Cloudflare’s 1.1.1.1 had Serve-Stale enabled, and their detailed incident report showed that users hitting warm caches continued to get working answers for .de domains — stale data, but functional. For most use cases (websites, APIs, mail routing), a stale A or AAAA record from five minutes ago is infinitely better than SERVFAIL.

The limitation: Serve-Stale only works if the record was previously cached. Cold caches — new queries for domains the resolver hadn’t seen recently — still failed. And once stale TTLs expired (typically capped at 1–3 days depending on implementation), even warm caches would stop serving.

Negative Trust Anchors (RFC 7646)

Negative Trust Anchors (NTAs) are the emergency brake for DNSSEC. An NTA tells a resolver: „Stop validating DNSSEC for this specific domain or zone.“ When applied to .de, it effectively disables signature verification for the entire TLD, allowing queries to resolve normally — at the cost of losing DNSSEC protection.

Cloudflare, Google, and Quad9 all deployed NTAs for .de within the first hour of the incident. This was the fastest path to restoring service for end users. The NTAs were removed once DENIC republished correct signatures later that day.

The Third Option: Disabling DNSSEC Validation Entirely

Some smaller operators chose the nuclear option: disabling DNSSEC validation on their resolvers entirely. This restored service for all domains immediately but removed cryptographic protection for every zone, not just the broken one. This is the equivalent of disabling your firewall because one rule is misconfigured — it works, but the security implications are severe. NTAs are strictly preferable because they scope the trust bypass to the affected zone.

The Amplification Problem

DNS outages create a vicious feedback loop. When resolvers return SERVFAIL, clients retry — aggressively. Applications retry. Browsers retry. Stub resolvers retry. Monitoring systems fire off their own queries. Cloudflare reported a 10x spike in query volume for .de during the incident, as retry storms amplified the load on authoritative servers and resolvers alike.

This client-retry amplification is a well-known pattern in distributed systems, but it’s especially brutal in DNS because retries happen at multiple layers simultaneously. It delays recovery because even after the root cause is fixed, the query flood continues until retry backoffs settle.

Parallels to Prior TLD Outages

The .de incident wasn’t the first time a TLD’s DNSSEC misconfiguration caused widespread outages. In 2024, New Zealand’s .nz experienced a similar DNSSEC signing failure that took down domains across the country. Sweden’s .se has had its own DNSSEC-related incidents. Each time, the pattern is the same: a key management error at the TLD level cascades into a nationwide or zone-wide outage, and the community rediscovers that DNSSEC’s strict validation model trades availability for integrity.

The lesson keeps repeating because the operational complexity of DNSSEC key management is genuinely hard, and the failure mode is binary: it either validates or it doesn’t. There’s no graceful degradation built into the protocol itself.

Platform Engineering Lessons

If you’re running a platform team — especially one operating in the EU — the .de incident should be a wake-up call. DNS is deeply embedded in every layer of a modern cloud-native stack: ExternalDNS syncs records, cert-manager validates domain ownership via DNS-01 challenges, Ingress controllers rely on DNS routing, service meshes resolve endpoints. A DNS outage isn’t just „websites are down“ — it can break certificate issuance, deployment pipelines, service discovery, and monitoring.

1. Monitor DNSSEC Validation, Not Just Resolution

Most teams monitor whether DNS resolution works. Few monitor whether DNSSEC validation is healthy. Set up checks that specifically test DNSSEC signature validity for your critical domains and their parent zones. Tools like DNSViz, Zonemaster, and RIPE Atlas probes can automate this. Alert on validation failures before your users notice.

2. Implement a Multi-Resolver Strategy

Don’t depend on a single upstream resolver. Configure failover across multiple providers: Cloudflare (1.1.1.1), Google (8.8.8.8), Quad9 (9.9.9.9). Each operator has different NTA deployment speeds and Serve-Stale configurations. During the .de incident, the window between „Cloudflare deployed NTA“ and „smaller ISP resolvers deployed NTA“ was measured in hours. A multi-resolver setup lets you ride the fastest responder.

3. Deploy Serve-Stale in Your Own Resolvers

If you run local resolvers (CoreDNS, Unbound, BIND), enable Serve-Stale. In CoreDNS, this means configuring the cache plugin with serve_stale. In Unbound, set serve-expired: yes with appropriate serve-expired-ttl and serve-expired-client-timeout values. This single configuration change is your best passive defense against upstream DNSSEC failures.

# Unbound example
server:
    serve-expired: yes
    serve-expired-ttl: 86400
    serve-expired-client-timeout: 1800
# CoreDNS example
.:53 {
    forward . 1.1.1.1 8.8.8.8 9.9.9.9
    cache 3600 {
        serve_stale 86400
    }
}

4. Treat DNS as a Critical Dependency in Your Architecture

Map out every component in your stack that depends on DNS resolution. ExternalDNS, cert-manager (DNS-01 challenges), Ingress controllers, external API calls, webhook endpoints, OAuth/OIDC provider discovery — all of these break when DNS breaks. Document these dependencies and include DNS failure scenarios in your chaos engineering practice.

5. Build a DNS Incident Response Playbook

Your runbook should5 include:

  • Detection: Automated alerts for DNSSEC validation failures and elevated SERVFAIL rates
  • Triage: Is the issue local, resolver-level, or TLD-level? Use dig +dnssec and delv to isolate
  • Mitigation: Pre-approved steps to deploy NTAs on local resolvers, switch upstream resolvers, or enable Serve-Stale
  • Communication: Templates for status page updates that explain DNS issues to non-technical stakeholders
  • Recovery: Validation that DNSSEC signatures are correct before removing NTAs

6. NIS2 and DORA: DNS Resilience Is Now a Compliance Issue

For organizations operating in the EU, the NIS2 Directive and the Digital Operational Resilience Act (DORA) explicitly require resilience measures for critical infrastructure, including ICT supply chain risks. DNS is a foundational ICT service. A TLD-level outage that takes down your platform because you had no failover, no Serve-Stale, and no incident playbook is now a compliance gap, not just an operational one. Document your DNS resilience measures as part of your NIS2/DORA risk assessments.

The Bigger Picture

The .de DNSSEC meltdown highlights a fundamental tension in internet infrastructure: the systems designed to protect us (DNSSEC, certificate validation, strict security policies) can also become single points of failure when they break. The answer isn’t to disable security — it’s to build resilience layers that absorb the impact of failures without sacrificing protection during normal operations.

Serve-Stale and Negative Trust Anchors are exactly this kind of resilience layer. They don’t weaken DNSSEC; they give operators a controlled way to maintain availability while the underlying issue is fixed. Every platform team should have both in their toolkit.

Conclusion: Your DNS Is Only as Strong as Your Weakest Trust Anchor

The .de outage wasn’t caused by a sophisticated attack. It was a configuration error during routine key rotation — the kind of mistake that can happen to any registry, any operator, at any time. What separated the teams that weathered it from those that scrambled was preparation: multi-resolver setups, Serve-Stale configurations, DNSSEC monitoring, and tested incident playbooks.

Your action items for this week:

  1. Check if your resolvers have Serve-Stale enabled. If not, enable it today.
  2. Set up DNSSEC validation monitoring for your critical domains and their parent TLDs.
  3. Document your DNS dependencies and add DNS failure to your incident response playbook.
  4. Test a multi-resolver failover — don’t wait for the next TLD outage to find out if it works.

The next DNSSEC meltdown isn’t a matter of if — it’s a matter of which TLD and when. Be ready.

Veröffentlicht am

Small Language Models for Platform Engineering: Why 8B Parameters Beat API Dependencies

The economics of AI in platform engineering are shifting — fast. For the past two years, the default answer to „how do we add AI to our internal platform?“ has been „call an API.“ But with inference costs rising, data governance getting stricter, and a new generation of compact models matching much larger counterparts on critical benchmarks, that default is worth questioning. Small Language Models (SLMs) — particularly in the 7B–9B parameter range — have reached a threshold where they can handle the majority of platform engineering workloads without ever leaving your network.

The Benchmark Reality Check: 8B Is Not a Compromise

IBM’s Granite 4.1 8B, released in April 2026 under Apache 2.0, is a useful anchor for this conversation. On enterprise coding benchmarks, the 8B model matches IBM’s own 32B Mixture-of-Experts (MoE) variant. On HumanEval pass@1, the 8B scores 87.2% compared to 89.6% for the 30B model — a gap of less than 3 percentage points that is largely irrelevant for the deterministic, constrained tasks that platform teams actually run.

This pattern holds across the SLM landscape:

  • Phi-4 (14B) — Microsoft’s model excels at reasoning-heavy tasks, punching well above its weight on MATH and GPQA
  • Qwen-3 (8B) — Strong multilingual coding support, excellent for polyglot infrastructure codebases
  • Llama-3.3 (8B) — Meta’s workhorse, widely supported across inference frameworks
  • Mistral-Small (22B) — A good middle ground when you need more capacity without the frontier price tag

The takeaway: if you are still reaching for GPT-4 or Claude Sonnet to answer „why is this Helm chart failing?“ you are likely overspending.

Dense Non-Thinking Architecture: Why It Matters for Operations

Granite 4.1 uses what IBM calls a Dense Non-Thinking Architecture. In practice, this means the model does not execute an internal chain-of-thought (CoT) reasoning step before responding. For frontier models solving novel math problems, CoT is valuable. For a platform engineer asking „summarize this PagerDuty alert and suggest the top three actions,“ CoT overhead is pure latency and token cost with zero benefit.

Platform tasks are largely pattern-matching with context, not novel reasoning. Alert triage, PR description generation, runbook execution, code review comments — these are well-defined, repetitive, structured tasks where a fast, confident response beats a slow, deeply deliberative one. Dense models optimized for inference speed are a natural fit.

The FinOps Case: What Self-Hosting an 8B Model Actually Costs

Let’s put numbers on this. A mid-tier platform team might generate 50,000 LLM calls per month for internal tooling: PR review summaries, alert enrichment, documentation queries, CI/CD pipeline diagnostics.

At $0.002 per 1K tokens (input + output average), 50,000 calls at ~500 tokens each = $50/month in API costs. Manageable — until agents arrive.

Agentic workflows are not single API calls. A single „investigate this alert“ agent might issue 15–25 tool calls, each with full context. That same 50,000-event scenario becomes 750,000–1,250,000 LLM calls. At $0.002/1K tokens, that is now $1,500–$2,500/month — and growing linearly with adoption.

Self-hosting an 8B model on a single RTX 4090 (~$1,800 hardware) or a Mac Studio M4 Max (~$2,000) delivers:

  • ~30–50 tokens/second throughput (sufficient for internal tooling)
  • Zero marginal cost per call after hardware amortization
  • Full data residency — no tokens leave your network
  • Instant availability without rate limits or provider outages

At an agentic scale, the hardware pays for itself within 1–2 months. Beyond that, it is pure savings.

Platform Engineering Use Cases Where SLMs Shine

1. Alert Triage and Runbook Execution

The HolmesGPT pattern (CNCF Sandbox) demonstrates the right approach: give an SLM access to kubectl, PromQL, and Loki, and a structured Markdown runbook. With a well-crafted runbook, tool calls per investigation drop from 16+ to 2–4. An 8B model running locally handles this at millisecond latency with no data leaving the cluster.

2. CI/CD Pipeline Assistance

PR description generation, test coverage summaries, changelog drafting — these are low-complexity, high-volume tasks. An SLM integrated directly into your CI/CD pipeline (via Ollama’s REST API or a vLLM endpoint) can run as a pipeline step without any external dependency. No API key rotation. No rate limiting during a big release crunch.

3. Code Review Comments

Automated first-pass code review — style enforcement, security pattern flagging, documentation gaps — is exactly the kind of task where an 8B model is sufficient. The model does not need to understand your entire business domain; it needs to apply consistent rules to code diffs. Fine-tuning on your internal codebase further improves relevance.

4. Documentation and Runbook Generation

Keeping runbooks current is a perennial platform team pain point. An SLM that can read infrastructure-as-code, observe recent incident patterns, and generate or update Markdown documentation solves a real operational problem — without requiring a cloud API call for every update.

Enterprise Trust: Granite’s Compliance Credentials

IBM Granite 4.1 ships with two features that matter disproportionately in regulated industries: Guardian Models and cryptographic signing.

Guardian Models are companion classifiers that can check model inputs and outputs for compliance — harmful content, PII exposure, prompt injection attempts. This is built into the model ecosystem, not bolted on afterward. For financial services or healthcare platform teams, this is a significant differentiator versus a generic open-source model.

The cryptographic signing (with ISO certification) means you can verify model provenance. In an era where supply chain security is central to platform governance (see SLSA, Sigstore, in-toto), being able to verify that the model running in your cluster is exactly the model IBM published is not a minor detail.

The Multi-Model Strategy: SLM + Cloud for 80/20 Coverage

The most practical approach is not „replace all cloud APIs with SLMs“ — it is to route intelligently:

  • ~80% of tasks → Local SLM: Alert triage, CI/CD assistance, doc generation, code review, runbook execution, structured queries against internal data
  • ~20% of tasks → Cloud frontier model: Novel architecture decisions, complex multi-step reasoning, tasks requiring broad world knowledge not captured in your fine-tuned model

This mirrors how mature platform teams already think about compute: use the right tool at the right cost tier. An internal platform that routes requests based on complexity signals (task type, token budget, confidence threshold) gives you both cost efficiency and capability headroom.

Getting Started: Self-Hosting in the Platform Engineering Stack

The barrier to running an 8B model is lower than most teams expect:

  • Ollama — Single-command model serving, REST API, model library with one-line pulls (ollama pull granite3.3:8b)
  • LM Studio — Desktop GUI for evaluation, good for initial benchmarking before committing to infrastructure
  • vLLM — Production-grade serving with OpenAI-compatible API, batching, and quantization support; the right choice for Kubernetes-native deployments

For Kubernetes, vLLM running as a Deployment with a GPU node selector and an HPA on request queue depth is a reasonable production starting point. Pair it with an OpenAI-compatible API shim and your existing LLM-integrated tooling requires zero code changes to switch endpoints.

The Connection to Agentic Infrastructure

The Agentic Compute Cliff is real: GitHub Copilot paused new signups in April 2026 due to capacity constraints, and multiple cloud providers are experiencing GPU shortages. As agentic workloads scale — where a single developer workflow might trigger hundreds of LLM calls per hour — dependency on cloud inference is a reliability and cost risk.

SLMs running on internal infrastructure are not just a cost play. They are a resilience play. Your internal platform keeps working when the cloud provider has an outage. Your agents are not rate-limited during a major incident response. Your data never transits a network boundary you do not control.

When 8B Is Not Enough

Intellectual honesty matters here. SLMs are not the answer for everything:

  • Novel architecture decisions requiring broad reasoning across domains
  • Complex multi-step debugging across large, unfamiliar codebases
  • Tasks requiring deep world knowledge beyond your training/fine-tuning window
  • High-stakes customer-facing generation where quality variance is unacceptable

The skill is in classification — building a platform that knows when to route locally and when to escalate to a frontier model. That routing logic, often just a simple task classifier, is itself a good candidate to run on a local SLM.

Conclusion: Make the Economics Argument

The conversation about SLMs in platform engineering is no longer theoretical. The benchmarks have arrived. The tooling (Ollama, vLLM, LM Studio) is mature. The hardware cost is justified within months at agentic scale. And the privacy and compliance benefits — data residency, Guardian Models, cryptographic provenance — increasingly matter as organizations bring AI deeper into their software delivery lifecycle.

The 8B parameter class is not a compromise. It is a deliberate choice that aligns cost, performance, privacy, and operational simplicity for the tasks that platform teams actually run. Start with one use case — alert triage is a natural first target — measure the results, and expand from there. The API dependency you are paying for today may be entirely optional.

Veröffentlicht am

The Vercel Breach Playbook: What Platform Teams Must Do When Their PaaS Provider Gets Compromised

Today — April 19, 2026 — Vercel disclosed a security incident involving unauthorized access to its internal systems. The breach has been linked to the ShinyHunters group, a threat actor known for targeting SaaS platforms via social engineering and vulnerability exploitation. Vercel says a „limited subset of customers“ was impacted and recommends reviewing environment variables — particularly urging use of their Sensitive Environment Variable feature.

If you’re a platform engineer running production workloads on Vercel, this is your signal to act. Not tomorrow. Now.

But this post isn’t just about Vercel. It’s about what every platform team should do when the infrastructure they trust gets compromised — because this has happened before, and it will happen again.

We’ve Been Here Before

The Vercel breach follows a pattern that platform teams should recognize by now:

  • CircleCI (January 2023) — An engineer’s laptop was compromised, giving attackers access to customer environment variables, tokens, and keys. CircleCI’s guidance was unambiguous: rotate every secret, immediately. Teams that delayed paid the price.
  • Codecov (April 2021) — Attackers modified Codecov’s Bash Uploader script, exfiltrating environment variables from CI pipelines for two months before detection. Thousands of repositories had their credentials silently harvested.
  • Travis CI (September 2021) — A vulnerability exposed secrets from public repositories, including signing keys and access tokens. The scope was enormous because the trust boundary had been quietly violated for years.

The common thread: environment variables are the crown jewels, and PaaS providers are the vault. When the vault gets cracked, every secret inside is potentially compromised.

The Shared Responsibility Blind Spot

Most teams understand the shared responsibility model for IaaS — you secure your workloads, AWS secures the hypervisor. But with PaaS providers like Vercel, Netlify, or Railway, the trust boundary is far murkier.

Consider what Vercel has access to in a typical deployment:

  • Your source code (pulled from Git during builds)
  • Every environment variable you’ve configured — database URLs, API keys, signing secrets
  • Build-time and runtime secrets
  • Deployment metadata and audit logs
  • DNS configuration and SSL certificates

When Vercel’s internal systems are breached, all of these become part of the blast radius. You didn’t misconfigure anything. You didn’t leak a credential. Your provider’s security posture became your security posture.

This is the platform trust boundary problem: the more convenience your PaaS offers, the more implicit trust you’ve delegated.

Immediate Response: The First 24 Hours

If you’re running on Vercel right now, here’s the checklist. Don’t wait for their investigation to conclude — assume the worst and work backward.

1. Audit Your Environment Variables

Vercel’s own advisory specifically calls out environment variables. Start here:

# List all Vercel projects and their env vars
vercel env ls --environment production
vercel env ls --environment preview
vercel env ls --environment development

Or use the consolidated environment variables page Vercel provides. Document every secret. You need to know what’s potentially exposed before you can rotate.

2. Rotate Every Secret — No Exceptions

This is the lesson from CircleCI: partial rotation is no rotation. If a secret was accessible to your PaaS provider, treat it as compromised.

  • Database credentials (connection strings, passwords)
  • API keys (Stripe, Twilio, SendGrid, any third-party service)
  • OAuth client secrets
  • JWT signing keys
  • Webhook secrets
  • Encryption keys

Prioritize by blast radius: payment processing keys and database credentials first, monitoring API keys last.

3. Review Deployment History

Check for unauthorized deployments or unexpected build activity:

# Review recent deployments via Vercel CLI
vercel ls --limit 50

# Check for deployments from unexpected branches or commits
vercel inspect <deployment-url>

Look for deployments that don’t correlate with your Git history. An attacker with access to Vercel’s internals could potentially trigger builds with modified environment variables or injected build steps.

4. Revoke and Regenerate Tokens

Beyond environment variables, rotate all integration tokens:

  • Vercel API tokens (personal and team)
  • Git integration tokens (GitHub/GitLab app installations)
  • Any webhook endpoints that use shared secrets for verification
  • CI/CD integration tokens that connect to Vercel

5. Check Downstream Systems

If your database credentials were in Vercel env vars, check your database audit logs for unusual access patterns. If your AWS keys were stored there, review CloudTrail. Every secret that was in Vercel is a thread to pull.

Stop Storing Secrets in Environment Variables

The deeper lesson here is architectural. Environment variables are the de facto standard for passing configuration to applications — but they were never designed as a secrets management system. They’re plaintext, they get logged, they get copied into build caches, and they’re only as secure as the system storing them.

External Secrets Operator

If you’re running Kubernetes workloads (even alongside a PaaS), the External Secrets Operator lets you reference secrets from external stores without ever putting them in your deployment platform:

apiVersion: external-secrets.io/v1beta1
kind: ExternalSecret
metadata:
  name: database-credentials
spec:
  refreshInterval: 1h
  secretStoreRef:
    name: vault-backend
    kind: ClusterSecretStore
  target:
    name: db-creds
  data:
    - secretKey: password
      remoteRef:
        key: secret/data/production/database
        property: password

The secret lives in Vault or AWS Secrets Manager. Your PaaS never sees it. If the PaaS is breached, the secret isn’t in the blast radius.

HashiCorp Vault with Dynamic Secrets

Even better: don’t store long-lived credentials at all. Vault’s dynamic secrets generate short-lived database credentials on demand:

# Application requests temporary database credentials at startup
vault read database/creds/my-role
# Returns credentials valid for 1 hour
# Automatically revoked after TTL expires

When your PaaS is breached, there’s nothing useful to steal — the credentials expired hours ago.

CI/CD Credential Hygiene: Kill the Static Tokens

Static API keys and long-lived tokens are the gift that keeps giving — to attackers. Every major PaaS breach has involved harvesting static credentials. The fix is structural.

OIDC Federation: Identity Without Secrets

Instead of storing cloud provider credentials in your CI/CD platform, use OIDC federation. Your pipeline proves its identity to the cloud provider directly, receiving short-lived tokens that can’t be stolen from the PaaS:

# GitHub Actions example — no AWS keys stored anywhere
- uses: aws-actions/configure-aws-credentials@v4
  with:
    role-to-assume: arn:aws:iam::123456789:role/deploy-role
    aws-region: eu-central-1
    # No access-key-id or secret-access-key needed
    # GitHub's OIDC token proves the workflow's identity

All major cloud providers support OIDC federation from GitHub Actions, GitLab CI, and most CI/CD platforms. There is no good reason to store static cloud credentials in your PaaS in 2026.

Workload Identity and SPIFFE7.

For more complex deployments, SPIFFE (Secure Production Identity Framework for Everyone) and its reference implementation SPIRE provide cryptographic identity attestation for workloads. Every workload gets a verifiable identity (SVID) without static credentials, and identity is attested based on the workload’s environment — not a secret that can be exfiltrated.

This is zero-trust for deployment pipelines: trust is established through verifiable identity, not shared secrets.

SBOM and Provenance: Know What You Shipped

When your build platform is compromised, one critical question emerges: can you prove that what’s running in production is what you intended to ship?

Build provenance — cryptographic attestations that link a deployed artifact to its source code, build parameters, and builder identity — becomes essential during incident response:

# Verify build provenance with cosign
cosign verify-attestation \
  --type slsaprovenance \
  --certificate-identity builder@your-org.iam.gserviceaccount.com \
  --certificate-oidc-issuer https://accounts.google.com \
  ghcr.io/your-org/your-app:latest

If you maintain SBOMs (Software Bills of Materials) and SLSA provenance attestations, you can forensically verify whether a compromised build platform injected anything into your artifacts. Without them, you’re flying blind.

Long-Term: Multi-Provider Resilience

The uncomfortable truth is that every PaaS provider will eventually have a security incident. The question isn’t if — it’s whether your architecture limits the blast radius when it happens.

Reduce Single Points of Trust

  • Secrets in an external vault, not in the PaaS — Vault, AWS Secrets Manager, Azure Key Vault
  • Build artifacts signed independently — don’t rely on the build platform’s integrity alone
  • DNS and TLS managed separately — if your PaaS controls your DNS, a breach can redirect traffic
  • Audit logs forwarded in real-time — ship PaaS audit logs to your own SIEM before the provider can tamper with them

Portable Deployments

If your deployment is tightly coupled to a single PaaS, you can’t move quickly during an incident. Containerized workloads with Infrastructure-as-Code configuration give you the option to shift to another platform within hours, not weeks. You don’t need to be multi-cloud on day one — but you need the capability to move when the trust relationship breaks.

The Incident Response Checklist

Pin this somewhere visible. When your next PaaS breach notification lands in your inbox:

Timeframe Action
0-1 hours Inventory all secrets stored in the provider. Begin rotating critical credentials (database, payment, auth).
1-4 hours Revoke all API tokens and integration credentials. Review deployment history for anomalies.
4-12 hours Complete rotation of all remaining secrets. Check downstream system audit logs. Verify build artifact integrity.
12-24 hours Confirm no unauthorized deployments occurred. Brief stakeholders. Document timeline.
1-7 days Conduct full post-incident review. Implement architectural improvements (external secrets, OIDC federation). Update runbooks.

Trust, but Architect for Betrayal

The Vercel breach is a reminder that platform trust is borrowed, not owned. Every convenience a PaaS provides — environment variable storage, built-in secrets, managed DNS — is a trust delegation that becomes a liability during a breach.

The platforms you depend on will get compromised. The question is whether you’ve architected your systems so that a provider breach is a inconvenience you handle in hours — or a catastrophe that takes weeks to untangle.

Start rotating your secrets now. Then start building the architecture that means you won’t have to do it so urgently next time.

Veröffentlicht am

Dapr Agents v1.0: Resilient Multi-Agent Orchestration on Kubernetes

The Distributed Systems Foundation for AI Agents

When LangGraph introduced stateful agents and CrewAI popularized role-based collaboration, they solved the what of multi-agent AI systems. But as organizations move from demos to production, a critical question emerges: how do you run these systems reliably at scale?

Enter Dapr Agents, which reached v1.0 GA in March 2026. Built on the battle-tested Dapr runtime—a CNCF graduated project—this Python framework takes a fundamentally different approach: instead of bolting reliability onto AI frameworks, it brings AI agents to proven distributed systems primitives.

The result? AI agents that inherit decades of distributed systems wisdom: durable execution, exactly-once semantics, automatic retries, and the ability to survive node failures without losing state.

Why Traditional Agent Frameworks Struggle in Production

Most AI agent frameworks were designed for prototyping. They work brilliantly in Jupyter notebooks but encounter friction when deployed to Kubernetes:

  • State Loss on Restart: LangGraph checkpoints require manual persistence configuration. A pod restart can lose agent memory mid-conversation.
  • No Native Retry Semantics: When an LLM API returns a 429, most frameworks fail or require custom retry logic.
  • Coordination Complexity: Multi-agent communication typically requires custom message queues or REST endpoints.
  • Observability Gaps: Tracing an agent’s reasoning across multiple tool calls often means stitching together fragmented logs.

Dapr Agents addresses each of these by standing on the shoulders of infrastructure patterns that have been production-hardened since the early days of microservices.

Architecture: Agents as Distributed Actors

At its core, Dapr Agents builds on three Dapr building blocks:

1. Workflows for Durable Execution

Every agent interaction—LLM calls, tool invocations, state updates—is persisted as a workflow step. If the agent crashes mid-reasoning, it resumes exactly where it left off:

from dapr_agents import DurableAgent, tool

class ResearchAgent(DurableAgent):
    @tool
    def search_arxiv(self, query: str) -> list:
        return arxiv_client.search(query)
    
    async def research(self, topic: str):
        papers = await self.search_arxiv(topic)
        summary = await self.llm.summarize(papers)
        return summary

Under the hood, Dapr Workflows use the Virtual Actor model—the same pattern that powers Orleans and Akka. Each agent is a stateful actor that can be deactivated when idle and reactivated on demand, enabling thousands of agents to run on a single node.

2. Pub/Sub for Event-Driven Coordination

Multi-agent systems need reliable communication. Dapr’s Pub/Sub abstraction lets agents publish events and subscribe to topics without knowing about the underlying message broker:

from dapr_agents import AgentRunner

await agent_a.publish("research-complete", {
    "topic": "quantum computing",
    "findings": summary
})

@runner.subscribe("research-complete")
async def handle_research(event):
    await writer_agent.draft_article(event["findings"])

Swap Redis for Kafka or RabbitMQ without changing agent code.

3. State Management for Agent Memory

Conversation history, tool results, reasoning traces—all flow through Dapr’s State API with pluggable backends:

from dapr_agents import memory

agent = ResearchAgent(memory=memory.InMemory())

agent = ResearchAgent(
    memory=memory.PostgreSQL(
        connection_string=os.environ["PG_CONN"],
        enable_vector_search=True
    )
)

Agentic Patterns Out of the Box

Dapr Agents ships with implementations of common multi-agent patterns:

Pattern Description Use Case
Prompt Chaining Sequential LLM calls where each output feeds the next Document processing
Evaluator-Optimizer One LLM generates, another critiques in a loop Code review
Parallelization Fan-out work to multiple agents, aggregate results Research synthesis
Routing Classify input and delegate to specialist agents Customer support
Orchestrator-Workers Central coordinator delegates subtasks dynamically Complex workflows

MCP and Cross-Framework Interoperability

A standout feature is native support for the Model Context Protocol (MCP):

from dapr_agents import MCPToolProvider

tools = MCPToolProvider("http://mcp-server:8080")
agent = DurableAgent(tools=[tools])

Dapr Agents can also invoke agents from other frameworks as tools:

from dapr_agents.interop import CrewAITool

research_crew = CrewAITool(crew=research_crew, name="research_team")
coordinator = DurableAgent(tools=[research_crew])

Kubernetes-Native Deployment

apiVersion: apps/v1
kind: Deployment
metadata:
  name: research-agent
  annotations:
    dapr.io/enabled: "true"
    dapr.io/app-id: "research-agent"
spec:
  replicas: 3
  template:
    spec:
      containers:
      - name: agent
        image: myregistry/research-agent:v1

Comparison: Dapr Agents vs. LangGraph vs. CrewAI

Capability Dapr Agents LangGraph CrewAI
Durable Execution Built-in Requires config Limited
Auto Retry Built-in Manual Manual
State Persistence 50+ backends SQLite, PG In-memory
Kubernetes Native Sidecar Manual Manual
Observability OpenTelemetry LangSmith Limited

When to Choose Dapr Agents

Dapr Agents makes sense when:

  • You’re already running Dapr for microservices
  • Your agents must survive node failures without state loss
  • You need to scale to thousands of concurrent agents
  • Enterprise observability requirements demand OpenTelemetry

Getting Started

pip install dapr-agents
dapr init

from dapr_agents import DurableAgent, AgentRunner

class GreeterAgent(DurableAgent):
    system_prompt = "You are a helpful assistant."

runner = AgentRunner(agent=GreeterAgent())
runner.start()

The Bigger Picture

Dapr Agents represents a broader trend: AI frameworks are maturing from „make it work“ to „make it work reliably.“ The CNCF ecosystem is converging on this need—KubeCon 2026 showcased kagent, AgentGateway, and the AI Gateway Working Group.

For platform teams, Dapr Agents offers a familiar operational model: sidecars, state stores, message brokers, and observability pipelines. The agents are new; the infrastructure patterns are proven.

Dapr Agents v1.0 is available now at github.com/dapr/dapr-agents.