Softlander Book a call
Agent sandbox assurance

Resource-exhaustion testing

Establish what happens when agent execution consumes excessive resources

Unbounded loops or large generated tasks affect neighboring workloads. The decision is how to establish what happens when agent execution consumes excessive resources. The work begins with your actual workflow and looks for the evidence that separates intended behaviour from a consequential exception. It is particularly relevant to SRE leads and shared-runtime owners. A useful outcome is an explicit account of what was examined, what remains uncertain, and which decision the evidence can support.

Book a 30 minute readiness callBrowse this capability group

The runtime isolation relationship

  1. Starting evidenceResource quotas

    Establish the basis for workload quota.

  2. Decision boundaryTermination path

    Observe throttling and forced-stop behaviour.

  3. Decision artefactResource-limit assessment

    Retain the scoped finding and its unresolved dependencies.

Proposed working model. Untested surfaces remain outside the conclusion.

Decision

Frame the decision around workload quota

Questions and the evidence they need

Criterion / evidenceWorkload quota
Inspect resource quotas for the relevant constraint or observation.
Criterion / evidenceTermination path
Inspect scheduler settings for the relevant constraint or observation.
Criterion / evidenceNeighbor workload
Inspect workload limits for the relevant constraint or observation.
Read each criterion against the available evidence before deciding that additional work is needed.

The decision is how to establish what happens when agent execution consumes excessive resources. This is relevant to SRE leads and shared-runtime owners when unbounded loops or large generated tasks affect neighboring workloads. Name the affected workflow, the consequence of a wrong decision, and the owner who can act on the result. The initial discussion should establish which evidence is already available and which uncertainty is material enough to investigate.

The specific failure hypothesis is that one task exhausts memory, disk, process count or execution slots before limits stop it. It remains a hypothesis until examined in the deployment. Existing internal tests or operating records may already resolve it; identify the uncovered question before commissioning additional work.

Mechanism

Resource limits should contain the effect on neighbors

Relationships within runtime isolation

Source or conditionWorkload quota
Bound CPU, memory, time, and output growth.
Crossing or choiceTermination path
Observe throttling and forced-stop behaviour.
Consequence or decisionNeighbor workload
Check interference and recovery after the test.
Compare effective controls with harmless test observations. Each relationship needs evidence in the actual scope.

What this specialty examines

Resource-exhaustion testing needs bounded workloads and clear stop conditions. Identify limits for CPU, memory, storage, process count, execution duration and shared scheduling where relevant. Observe neighboring workloads so a task’s own termination does not conceal wider disruption.

The review should run bounded synthetic stress cases and observe throttling, termination and neighboring task behaviour. Identify the normal route, relevant exceptions and the authoritative record of the outcome. Inspect the relevant business-system record, enforcement event or process evidence alongside any reported status. Keep incomplete cases visible, with an explanation of what their absence prevents the owner from concluding.

The tradeoff to make explicit

For this decision, strict limits protect shared capacity while legitimate long-running work needs an explicit route. Compare the available design or operating choices against the same intended outcome. If one option removes a required control or changes the task boundary, describe that difference before treating its apparent simplicity, speed or lower effort as an improvement. The decision should expose the tradeoff instead of burying it in an aggregate score.

Preparation

Connect the working inputs to the boundary being examined

Evidence packet for workload quota

Illustrative working format

BasisResource quotas
Establish permitted access, provenance, and which task conditions it represents.
ContextScheduler settings
Check its relationship to the relevant workflow and expose missing context.
ObservationWorkload limits
Keep workload, quality, human effort, and uncertain inputs visible in comparisons.
DependencyTelemetry
Identify provenance and the specific decision this material can inform.
Illustrative preparation record. Availability, revision, and handling need confirmation.

Start with resource quotas, scheduler settings, workload limits and telemetry. Identify the owner, revision and operating context of each source. Sanitised examples may support initial design, but differences from actual conditions should remain explicit. Where direct access is unsuitable, a customer-controlled evidence or execution route may be considered, with agreed handling and retention.

The scope also needs the runtime owner, agreed expectations for a representative boundary test, and the authority to inspect the relevant boundary. Specify what evidence would distinguish the failure in which one task exhausts memory, disk, process count or execution slots before limits stop it from acceptable behaviour. If stakeholders disagree about intended behaviour, resolve the interpretation before classifying the observed difference as a defect.

How the work proceeds

Carry workload quota through to an accountable finding

From the scoped material to a decision record

  1. Establish conditionsResource quotas

    Declare the permitted environment and resources.

  2. Examine the relationshipTermination path

    Apply the scoped check to neighbor workload.

  3. Retain the findingResource-limit assessment

    Record the evidence, open dependencies, and the owner’s next decision.

Proposed sequence; resolve interpretation or access gaps before relying on the next step.

Prepare the runtime boundary and fixtures

Use resource quotas, scheduler settings, workload limits and telemetry to identify the exact runtime and boundary claimed by SRE leads and shared-runtime owners. Mark permitted resources and protected resources in a controlled fixture, including shared services where they are relevant. Define observable success, denial and incomplete setup before execution. The preparation output is a boundary map and fixture manifest that lets reviewers distinguish effective isolation from a test that never reached its target.

Exercise and inspect the boundary

Distinguish throttling, graceful cancellation and abrupt failure. Each has implications for partial artefacts, retries and cleanup. Record whether limits are enforced per process, session, tenant or shared pool, since those scopes produce different operating behaviour.

Choose containment or configuration changes

Translate the observations into a resource-limit assessment with termination and recovery findings. Locate each issue in the worker, shared service, policy rollout or cleanup path rather than attributing every symptom to the sandbox as a whole. Decide which configuration change, narrower resource exposure or specialist investigation is warranted. Retest the affected boundary and inspect residual state before treating the changed environment as equivalent to the intended design.

Indicative artefact

Make the resource-limit assessment usable by its owner

Resource-limit assessment

Illustrative working format

Annotated fieldResource-limit assessment
Identify the assessed workflow, revision, and decision boundary.
Annotated fieldTermination
Link supporting and contrary observations to inspectable evidence.
Annotated fieldRecovery findings
Connect the triggering state to permitted action and evidence of recovered business state.
Illustrative artefact structure, not a completed engagement or an observed result.

A useful decision artefact is a resource-limit assessment with termination and recovery findings. Supporting material may include selected cases, configuration or process records, evidence references, a disagreement log and prioritised actions. Retain the case selection rationale and material disagreements alongside the result; otherwise the result may be mistaken for complete coverage.

Illustrative scenario

Examine neighbor workload through a constructed exception

A constructed case and its decision paths

Illustrative scenario

  1. SituationWorkload quota

    A bounded stress task is stopped while another tenant's task continues.

  2. Question to examineTermination path

    Observe quota response and neighboring workload behaviour during bounded stress.

  3. Conditional next actionNeighbor workload

    Retain termination and recovery evidence alongside the configured limit.

Fictional example. No executed test, customer engagement, or measured outcome is implied.

Consider a fictional situation: a code agent generates a rapidly growing temporary file that fills shared storage for unrelated sessions. The scenario is illustrative, with no client engagement, executed test or measured improvement implied.

For the growing temporary file, cap the synthetic test before shared infrastructure is harmed. Inspect quota enforcement and recovery after termination. The useful decision concerns containment and cleanup behaviour, not achieving the largest possible resource spike.

Scope and next action

Keep the conclusion within the authorised runtime boundary

Confirm fit before the next commitment

Included boundaryWorkload quota
This review is limited to the authorised runtime boundary.
PrerequisiteResource quotas
The relevant owner must confirm access and the intended operating conditions.
Claim limitResource-limit assessment
Untested surfaces remain outside the conclusion.
Bring the triggering change and existing evidence; resolve the dependency that could alter the decision.

Testing is confined to explicitly authorised environments and non-destructive cases. A bounded assessment cannot establish that escape or cross-boundary access is impossible. Any engagement depends on an agreed scope, qualified team, authorised access and confirmed availability. Bring the affected workflow, its current configuration or process, the triggering change and the evidence already available. The immediate scope question is prompted by the situation in which unbounded loops or large generated tasks affect neighboring workloads; confirm which missing observation would change the owner’s next action.

Ways to engage

Related engagement scopes

AI Agent Runtime Security Review

Make every delegation explicit. Examine who the agent acts for, where it can execute and which tools it can reach when instructions alone are not enough.

Read about this offer

AI Agent Red Team

Challenge the boundaries that matter to your workflow. Test agreed attack scenarios against agent instructions, tools and data access.

Read about this offer

AI Agent Sandbox Assurance

Examine the boundary between agent execution and the systems around it, including files, networks, secrets and cleanup.

Read about this offer

Continue exploring

Related insights and resources

Agent security6 MIN READ

The permission boundary behind every agent action

An agent can only be evaluated in the context of what its tools allow. Examine identity, delegated authority and action boundaries before you put a workflow in motion.

Whitepaper5 MIN READ

Before your agent can act, define its authority

A working paper for the moment an AI system moves from advice to action. Map identity, permissions, approvals and recovery into an authority model your team can review and test.

A little more clarity

Questions to consider.

Start with the questions that shape a better decision.

Which resources should be limited beyond CPU and memory?

Consider storage, process count, execution duration, open connections and shared scheduling in addition to CPU and memory. Select limits according to the workload and observe termination, cleanup and neighboring task effects.

Which quotas define acceptable execution resources?

Start with resource quotas, scheduler settings, workload limits and telemetry. Record their revision, owner and relationship to the intended workflow. Missing context should be identified before interpreting outcomes; a sanitised fixture is useful only when its relevant differences from the operating case are understood.

How should termination and neighboring workload effects be tested?

Run bounded synthetic stress cases and observe throttling, termination and neighboring task behaviour. Retain effective runtime configuration and observations from the relevant boundary. Distinguish an enforced denial from failed setup or an operation that never reached the resource. Use controlled markers and preserve residual-state evidence.

What belongs in a resource-limit assessment?

The proposed output is a resource-limit assessment with termination and recovery findings. Include the selected conditions, supporting observations, unresolved differences and next decision. A reviewer should be able to follow each important conclusion back to its evidence and understand the limits without depending on an undocumented explanation from the author.

When should a workload be throttled or stopped?

Throttle or stop when agreed resource limits are reached or shared-service protection requires it. Define cleanup and partial-work handling so termination does not leave persistent damage or uncontrolled retries.

What if existing evidence already answers the question?

Review its scope, provenance and configuration before proposing new work. Existing work from SRE leads and shared-runtime owners may already be sufficient. In that case, capture the decision, unresolved limitations and the change that would require another review. Additional assessment should address a real gap.

What would make the findings inconclusive?

Missing resource quotas, scheduler settings, workload limits and telemetry can prevent a reliable interpretation, depending on the case. Record the specific missing observation and its implication. The owner can supply evidence, narrow the claim or defer the dependent decision; an incomplete case should not silently be counted as a pass.

Start with the decision you need to make.

Tell us about the agent workflow you want reviewed, and where you need greater clarity. Scope and availability are confirmed together.

Book a 30 minute readiness call Get the agent release checklist

Connected disciplines

Malicious-code execution testing

Validate controls around potentially harmful generated code. The review connects execution policies, allowed interpreters, network limits and harmless test payloads to a bounded review of execution boundaries, giving the runtime owner a clearer basis for the next action.

Persistence and cleanup verification

Verify that finished agent sessions leave no unintended reusable state. The review connects session lifecycle events, volume policies, artefact lists and synthetic state markers to a bounded review of execution boundaries, giving the runtime owner a clearer basis for the next action.

Tenant-isolation assessment

Assess separation between tenants sharing agent execution infrastructure. The review connects tenant allocation rules, storage namespaces, cache keys and synthetic tenant cases to a bounded review of execution boundaries, giving the runtime owner a clearer basis for the next action.