ACKUITY BLOG

Kill switch or control switch?

Stop one unsafe action, or contain the agent? The right response depends on what has gone wrong and whether the rest of the task can still be trusted.

WATCH THE EXPLANATION

Is a kill switch enough?

A kill switch can stop or quarantine an unsafe agent. Per-action control adds selective decisions when authorized work can safely continue.

Chapters, full transcript, sources and answers

Consider a hypothetical support agent processing a queue of refunds. One request points to an account that does not belong to the customer. Can you stop that request while keeping legitimate work available? Or does the session show a wider compromise that requires containment?

A control should support the response your situation needs. The label alone does not tell you whether it acts on a tool, an action, one agent or the whole workflow.

SCOPE

What does each control actually stop?

Compare the intervention and its recovery requirements.
DecisionWhole-agent containmentPer-action control
What is stopped?The agent's continued operation or access, within the control's coverage.A proposed action before its effect reaches the target.
When does it fit?Continued execution cannot be trusted.One action needs intervention and the remaining task can safely continue.
What happens to other work?Queued and in-flight work may be interrupted. Recovery depends on checkpoints and downstream state.Authorized work can continue after the task and session are assessed as trustworthy.
What evidence is required?Proof of containment across the agent, outstanding work and relevant credentials.A verdict linked to the action and proof of the actual target-system outcome.

Some kill-switch implementations already provide capability-level or agent-level controls. Compare supported behavior and coverage rather than treating every alternative as an all-or-nothing shutdown. For definitions and implementation requirements, start with the AI agent kill-switch guide.

ONE ACTION AT A TIME

When should you block an action instead of the agent?

The same five agent actions under two kinds of switch. The sequence is db.read order history, api.post refund #400, api.post refund #401 to the wrong account, email.send refund receipt, and crm.update close ticket. Step 3 is the rogue action. With a kill switch, steps 1 and 2 run, the whole agent is quarantined at step 3, and the good work in steps 4 and 5 stops too. With a control switch, only step 3 is blocked, step 4 is constrained with personal data masked, and steps 1, 2 and 5 run. Terminate stays available for catastrophic cases.KILL SWITCHStop the agent1db.readorder historyRUNS2api.postrefund #400RUNS3api.postrefund #401, wrong accountQUARANTINED4email.sendrefund receiptSTOPPED5crm.updateclose ticketSTOPPED1 bad action stopped, 2 good ones lostCONTROL SWITCHStop the action1db.readorder historyRUNS2api.postrefund #400RUNS3api.postrefund #401, wrong accountBLOCKED4email.sendrefund receiptCONSTRAINED5crm.updateclose ticketRUNS1 blocked, 1 constrained, 3 runTERMINATE STAYS AVAILABLE FOR CATASTROPHIC CASES
This example contrasts whole-agent quarantine with per-action control. Quarantine can interrupt queued work; selective blocking preserves other work only when the session remains trustworthy. Kill-switch scope varies by implementation. Ackuity retains Terminate for catastrophic cases.

In the refund example, block the wrong destination before the payment runs. Preserve the case context and investigate why the request changed. A blocked action does not, by itself, establish that the next action is safe.

If the evidence supports a local mistake and the task remains trustworthy, a corrected proposal can receive a new decision. If the goal is hijacked, credentials are compromised or the agent is trying alternate paths around controls, contain the agent and investigate before resuming.

When I discuss the first case with our team, I call a whole-agent shutdown "overkill" for one bad move. The qualification matters: use the narrow response only when it contains the risk.

The refund is an illustrative scenario, not a reported customer incident or a completed product test.

DECISION CONTEXT

What makes Ackuity's control switch different?

Ackuity sits beside the agent in the execution path, outside the agent's control. It checks the action against the Agent Security Context Graph: 29 signals across the user, agent, intent and goal, target system and data, tools and supply chain, and history.

Deterministic policy checks enforce hard boundaries. Small language models interpret security context alongside rules and behavioral baselines. Within authorized scope, intent, history and risk determine execution or human review. No LLM judges the action.

A reviewer's approval cannot expand the agent's permissions. An authorized but unfamiliar bulk operation can require review; an operation that violates a hard policy limit remains blocked. Vinod's same-action context-graph example shows that distinction.

Ackuity checks 60+ threat models in 14 categories, mapped to NIST, OWASP and MITRE ATLAS. Decisions take 40 to 100 ms. That figure is decision time, separate from end-to-end overhead.

PROPORTIONATE RESPONSE

How does the response ladder work?

  1. Allow: signed, logged and run.
  2. Constrain: masked or scoped, then run.
  3. Human in the loop: paused for a person's decision.
  4. Block: dropped before the target.
  5. Terminate: container shut down, for catastrophic cases.

After a Block, do not silently retry the same action. After human review, changes to the target, amount or relevant context need a fresh decision. After Terminate, recovery requires a trusted execution state and an assessment of the affected credentials and downstream work.

COVERAGE

Which paths can the control enforce?

A control can enforce only the paths it covers. In Ackuity's routing model, 8 of 11 action types never reach a gateway. Your actual routing and credentials determine which boundary can stop a direct database call, shell command, queued task or agent hand-off.

The control belongs in the Goldilocks zone: not inside the agent, not at the gateway. Validate the boundary against the paths your agents actually use. The pre-execution implementation checklist explains what to check.

NVIDIA's Open Agent Safety Platform illustrates independent containment: OpenShell provides sandbox policy and NVIDIA describes Sentry as an out-of-band watchdog that can quarantine unsafe execution. Ackuity is an NVIDIA Inception member; our OpenShell supervisor integration is in development. Membership does not establish integration availability.

EVALUATE THE OUTCOME

What should a vendor demonstrate?

Use an isolated environment to observe an authorized action, an unauthorized action, a human-review boundary and whole-agent containment. Check the target-system record as well as the decision log. Then exercise alternate routes, queued work, child agents and a decision-service timeout.

Credential revoked, but work still running? Use the credential-revocation evaluation plan to separate a denied future request from an accepted job and a child agent’s own credentials.

Record supported paths, policy and product versions, legitimate-task completion, false positives, decision time and end-to-end overhead. The blank control-evaluation worksheet helps record those observations. It is a test plan, not evidence that a product has passed.

OWASP's Excessive Agency guidance recommends limiting privileges, requiring approval for consequential actions and enforcing authorization downstream. Use those principles alongside your containment and recovery tests.

Start with the 30-day Agent Exposure Assessment to examine existing telemetry and identify exposure. Inline enforcement and recovery behavior need a separate controlled evaluation.

NVIDIA, OpenShell and BlueField are trademarks of NVIDIA Corporation.

Putting human review into an execution path? Use the scoped-approval runbook and blank review record to define the operation, expiry, execution recheck and recovery evidence.

QUESTIONS PEOPLE ASK

Frequently asked questions

What is the difference between a kill switch and a control switch?

A kill switch contains unsafe execution, with scope that varies by implementation. Ackuity's Agent Execution Control Switch evaluates each proposed action before it runs and retains whole-agent termination for catastrophic cases. Compare actual scope, coverage and recovery behavior.

Can a kill switch target one agent without stopping everything else?

Some implementations can stop one agent or disable a capability. Check dependent agents, queued work and credentials. Per-action control provides a narrower decision when one proposed action is unsafe and the task can still be trusted.

When should you terminate an agent?

Contain or terminate the agent when continued execution cannot be trusted or individual action blocking cannot contain the risk. Investigate compromised goals, credentials, memory and bypass attempts before resuming.

Does blocking an action make later actions safe?

No. Investigate the reason for the block and assess whether the agent and task remain trustworthy. Corrected proposals need a fresh decision; a compromised session needs containment and investigation.

Can approval override a hard policy limit?

No. Hard policy boundaries remain enforced. Human review handles ambiguous actions within authorized scope, with approval tied to the specific action and context.

How do you test the difference?

Observe authorized and denied actions, verify the downstream outcome, test approval boundaries and termination, and inspect alternate routes, queued work, child agents and failure behavior. Record versions, policies, legitimate-task outcomes and latency.

KEEP READING

Keep exploring.

NEXT STEP

What does this mean for your agents?

Connect the ideas to your own tools, data and execution environment.

Request assessment