Softlander Book a call
Agent security and cybersecurity

Secure agent architecture assessment

Assess where security controls should sit in an agent architecture

The deployment adds tools, memory or privileged workflow steps. The decision is how to assess where security controls should sit in an agent architecture. 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 application security and platform architects. 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 agent security relationship

  1. Starting evidenceComponent diagrams

    Establish the basis for untrusted boundary.

  2. Decision boundaryDecision point

    Locate policy enforcement before action.

  3. Decision artefactArchitecture assessment

    Retain the scoped finding and its unresolved dependencies.

Proposed working model. A bounded challenge cannot establish exhaustive protection.

Decision

Frame the decision around untrusted boundary

Questions and the evidence they need

Criterion / evidenceUntrusted boundary
Inspect component diagrams for the relevant constraint or observation.
Criterion / evidenceDecision point
Inspect identity flows for the relevant constraint or observation.
Criterion / evidenceFailure behaviour
Inspect policy enforcement 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 assess where security controls should sit in an agent architecture. This is relevant to application security and platform architects when the deployment adds tools, memory or privileged workflow steps. 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 the architecture relies on prompt wording to enforce a boundary that tools do not check. 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

Control placement should follow the authority transition

Relationships within agent security

Source or conditionUntrusted boundary
Separate external input from trusted configuration.
Crossing or choiceDecision point
Locate policy enforcement before action.
Consequence or decisionFailure behaviour
Keep denial and dependency failure within scope.
Observe the control and resulting business effect. Each relationship needs evidence in the actual scope.

What this specialty examines

Architecture assessment examines where controls are enforced and what happens if one component behaves unexpectedly. Identify trust transitions between inputs, model, memory, orchestration, tools and business systems. A clear boundary should not depend entirely on cooperative model behaviour.

The review should review trust transitions, control placement, failure behaviour and separation of authority. 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, central enforcement simplifies policy consistency while distributed components need their own explicit guarantees. 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 untrusted boundary

Illustrative working format

BasisComponent diagrams
Match this description to the deployed or proposed boundary being examined.
ContextIdentity flows
Use nonsecret metadata to trace issuance, acceptance, and withdrawal responsibility.
ObservationPolicy enforcement
Confirm the applicable revision and who resolves disputed interpretation.
DependencyData paths
Establish permitted access, provenance, and which task conditions it represents.
Illustrative preparation record. Availability, revision, and handling need confirmation.

Start with component diagrams, identity flows, policy enforcement and data paths. 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 security decision owner, agreed expectations for a representative threat scenario, and the authority to inspect the relevant boundary. Specify what evidence would distinguish the failure in which the architecture relies on prompt wording to enforce a boundary that tools do not check 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 untrusted boundary through to an accountable finding

From the scoped material to a decision record

  1. Establish conditionsComponent diagrams

    Define the hypothetical attacker’s starting influence.

  2. Examine the relationshipDecision point

    Apply the scoped check to failure behaviour.

  3. Retain the findingArchitecture 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.

Define the controlled threat scenario

Review component diagrams, identity flows, policy enforcement and data paths with application security and platform architects. State what the hypothetical attacker controls, the protected action and the systems included in the authorised exercise. Prepare synthetic records and harmless markers where appropriate, with stopping conditions for unexpected effects. The preparation output is a scenario brief that distinguishes an influence attempt from a prohibited action actually reaching a business resource.

Challenge the path and review effects

Review control placement, denial behaviour and separation of duties. Check whether a compromised or mistaken planner can bypass the component meant to constrain it. Recommendations should explain which action a design change would prevent or make reviewable.

Select a control response and retest

Build an architecture assessment with control placement findings and proposed design options from the reviewed observations. Separate agent behaviour, application enforcement and the resulting resource state so the recommended change reaches the responsible layer. Choose a bounded retest that reproduces the original condition and checks legitimate behaviour after remediation. If the required investigation exceeds the agreed technique or expertise, retain the unresolved path and identify the qualified review needed.

Indicative artefact

Make the architecture assessment usable by its owner

Architecture assessment

Illustrative working format

Annotated fieldArchitecture assessment
Identify the assessed workflow, revision, and decision boundary.
Annotated fieldControl placement findings
Link observations and counterevidence to the case, configuration, and conclusion limit.
Annotated fieldDesign options
Name the unresolved condition, next action, and accountable owner.
Illustrative artefact structure, not a completed engagement or an observed result.

A useful decision artefact is an architecture assessment with control placement findings and proposed design options. 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 failure behaviour through a constructed exception

A constructed case and its decision paths

Illustrative scenario

  1. SituationUntrusted boundary

    An architecture relies on prompt wording where a tool-side decision is needed.

  2. Question to examineDecision point

    Locate the resource-side decision for the proposed operation.

  3. Conditional next actionFailure behaviour

    Move or strengthen enforcement where prompt interpretation cannot settle authority.

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

Consider a fictional situation: a planner receives unrestricted tool access and is expected to follow a natural-language rule limiting customer updates. The scenario is illustrative, with no client engagement, executed test or measured improvement implied.

For the unrestricted planner, consider tools that enforce action and resource scope independently. The planner can still propose work, while the executor applies an explicit policy. The architecture decision concerns enforceable authority rather than stronger instructional wording alone.

Scope and next action

Keep the conclusion within the agreed threat scenario

Confirm fit before the next commitment

Included boundaryUntrusted boundary
This review is limited to the agreed threat scenario.
PrerequisiteComponent diagrams
The relevant owner must confirm access and the intended operating conditions.
Claim limitArchitecture assessment
A bounded challenge cannot establish exhaustive protection.
Bring the triggering change and existing evidence; resolve the dependency that could alter the decision.

Authorised scenario testing is not a claim of exhaustive penetration testing or guaranteed protection. Intrusive security work requires qualified specialists and a separately agreed scope. 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 the deployment adds tools, memory or privileged workflow steps; confirm which missing observation would change the owner’s next action.

Ways to engage

Related engagement scopes

Agent Release Readiness Review

The Agent Release Readiness Review is a fixed-scope, three-week independent test of one AI agent workflow that shows a release board what the agent does, what it is allowed to do and what happens when it is wrong, and ends in a signed recommendation.

3 weeksRead about this offer

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

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.

Agent security6 MIN READ

Running an AI-agent incident tabletop exercise

A useful tabletop follows one plausible agent failure through detection, containment, evidence, business recovery and accountable return to service.

A little more clarity

Questions to consider.

Start with the questions that shape a better decision.

Which controls belong outside the prompt?

Place enforceable authorisation, parameter validation and resource restrictions at the components that execute actions. Prompts can express expectations, but the architecture should examine behaviour if the planner ignores or misunderstands them.

Which architecture records locate security controls?

Start with component diagrams, identity flows, policy enforcement and data paths. 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 bypass paths and denial behaviour be reviewed?

Review trust transitions, control placement, failure behaviour and separation of authority. State the hypothetical attacker's control and inspect attempted actions, tool enforcement and actual effects separately. A refusal, a blocked call and a successful prohibited action are different results. Preserve unsuccessful challenges as well as adverse findings.

What belongs in a control-placement assessment?

The proposed output is an architecture assessment with control placement findings and proposed design options. 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 enforcement move outside the planner?

Move enforcement outside the planner when the protected boundary must hold even if the agent misunderstands or ignores instructions. Apply authorisation and validation where actions execute, then test denied and permitted behaviour.

What if existing evidence already answers the question?

Review its scope, provenance and configuration before proposing new work. Existing work from application security and platform architects 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 component diagrams, identity flows, policy enforcement and data paths 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

Prompt-injection testing

Test whether direct instructions can redirect an agent beyond its approved task. The review connects instruction hierarchy, protected behaviours, synthetic prompts and tool audit records to a bounded review of agent security, giving the security decision owner a clearer basis for the next action.

Indirect prompt-injection assessment

Examine instructions hidden inside information an agent consumes. The review connects content provenance, retrieval paths, trust labels and consequential action rules to a bounded review of agent security, giving the security decision owner a clearer basis for the next action.

Tool and capability abuse testing

Understand whether legitimate tools can be combined for unintended actions. The review connects tool inventories, permitted use cases, action sequences and resource constraints to a bounded review of agent security, giving the security decision owner a clearer basis for the next action.