Policy · critical
HIGH risk - Destructive operations require human approval
EU AI Act Article 14(4)(d) requires that humans can decide not to use the AI system or override its output. DELETE operations are irreversible and require explicit human approval via a step.gate.
The policy
A few lines of YAML in your repository.
Part of EU AI Act, high risk. Assign the pack, or copy this policy into a manifest of your own.
- Verdict
- Critical: the step is blocked before it runs, and an incident is opened.
- Checked
- Before every step, against everything the agent did earlier in the task.
- Rule
- step_requires_gate
- name: "HIGH risk - Destructive operations require human approval" description: | EU AI Act Article 14(4)(d) requires that humans can decide not to use the AI system or override its output. DELETE operations are irreversible and require explicit human approval via a step.gate. Applies to: HIGH risk agents only. rule_type: step_requires_gate params: target_kinds: ["step.resource"] target_verb: DELETE gate_kind: human severity: critical enforcement_point: step_execution risk_classification: highNot legal advice. This policy is Kyvvu's interpretation of the EU AI Act. It isn't reviewed or endorsed by a regulator, and using it doesn't on its own make you compliant: it gives you technical controls, and a record that they ran, to support your compliance work.
Why it matters
Applies to: HIGH risk agents only.
How the rule works
step_requires_gate: Checks that a matching, completed step.gate precedes this step. By default the gate may be anywhere in history. Set max_steps_since_gate to require it among the last N entries; blocked, failed and in-flight attempts after the gate count toward N. A blocked gate cannot approve the target.
target_kinds:["step.resource"]. Action kinds the rule applies to (e.g. step.resource, step.exec, step.message); other steps pass.target_verb:DELETE. Only apply to target steps with this verb (e.g. GET, POST, DELETE).gate_kind:human. Only gates with this properties.kind count.
In these packs
- EU AI Act, high risk, version 1.0.2
From Kyvvu's open policy library at github.com/Kyvvu/manifests, Apache 2.0.
Next
Watch it decide.
In your browser
Try it in the playground
The policy is loaded. Add the steps your agent takes and watch the engine decide, step by step.
On your machine
Five minutes, no account
kyvvu try runs a demo agent against the bundled baseline. Then assign this pack from the dashboard or the CLI.
For security teams
See it on your agents
A 30-minute demo on the agents you actually run, with the policies that fit them.