Return an Approval-Required Signal to a Host-Owned Review Workflow
Hold payments, refunds, beneficiary changes, account restrictions, and other consequential mutations unless declared authority and policy permit them.
What this path changes
The host can block prohibited transactions, hold uncertain actions for an authorized reviewer, and keep the policy verdict separate from the reviewer decision.
The action, boundary, and evidence
Who owns the risk
Financial-services engineers, governance owners, operations teams, and product-risk leaders responsible for transaction authority.
When the gate runs
An agent proposes a payment, refund, transfer, beneficiary change, payout update, account restriction, or account-record mutation.
How to configure assurance
Use deterministic transaction, amount, destination, and authority rules. Add semantic observation for contextual anomalies. Missing required evidence or observer disagreement holds the action for policy-defined review.
- Send funds, issue a refund, or raise a transaction limit.
- Change beneficiary or payout details.
- Restrict, close, or materially modify an account.
- A policy test matrix covering permit, review, explicit block, missing evidence, and disagreement.
- A replayable decision record that keeps policy, reviewer choice, execution attempt, and destination outcome separate.
Next: Review an example payment action before connecting a financial tool.
Follow a matching proposed action through the gate.
Synthetic tested action under the pinned 5D 0.7.0 illustrative policy.
Issue a high-value refund to a stored payment method
A fabricated complaint or mistaken account match triggers value movement. Under the pinned illustrative policy, the expected result is to hold the action until the host records the required review.
- Tool
- refund.issue
- Target
- stored payment method
- Purpose
- resolve dispute
5D suppliesAn ORANGE band and hold signal.
Host ownsReviewer identity, authority, approval freshness, and final execution.
Coverage limit5D does not supply the approval queue or authenticate the reviewer.
Where this control fits in the governance lifecycle.
These references identify possible technical-evidence contributions. They do not mean every framework applies to every deployment.
This mapping identifies possible technical-evidence contributions. It is not legal advice, certification, conformity assessment, readiness scoring, a compliance conclusion, or proof that an AI deployment is safe. Applicability, governance, integration coverage, and deployment outcomes remain the organisation's responsibility.
Enforce at action time
May this proposed action cross the boundary?
Articles 9(5)–(6), 14 and 15(1), 15(4)Risk treatment, human oversight, robustness and cybersecurity
What 5D helps with: Pre-action prohibitions, routing and a policy-selected human hold can implement one technical risk-treatment measure.
What 5D does not establish: Only relevant where these duties apply to the actor and system; 5D does not establish the wider risk system, oversight design, robustness or cybersecurity.
Official source ↗MANAGE 1.2–1.3Prioritise and document risk treatment
What 5D helps with: Permit, hold and block policy can implement one documented treatment for selected action risks.
What 5D does not establish: 5D does not prioritise organisation-wide risk or determine risk acceptance and treatment sufficiency.
Official source ↗Observe and review
What happened, and can the decision be reconstructed and challenged?
Section 2 — Transparency and explainability; accountability and governance; contestability and redressTraceable decisions, accountability and challenge
What 5D helps with: Reconstructable action records can support internal explanation and challenge of one policy decision.
What 5D does not establish: 5D does not create affected-person notices, complaint channels, remedies, complete system explanations or regulator-specific accountability.
Official source ↗Dimensions 2 and 3 — Enable meaningful human accountability; implement technical controls and processesHuman checkpoints, logging and monitoring
What 5D helps with: A policy-selected hold and action receipt can support one human checkpoint and record the resulting disposition.
What 5D does not establish: The organisation owns reviewer identity, authority, competence, escalation, monitoring coverage and accountability.
Official source ↗Inspect the routing signal
from fivedrisk import classify_tool_call, score, load_policy
policy = load_policy("policy.yaml")
action = classify_tool_call("Bash", {"command": "deploy"}, policy)
result = score(action, policy)
print(result.routing.approval_required)Written for fivedrisk==0.7.0. Verify against the immutable source and public documentation before use.
Let policy select the review queue
The band and routing result indicate whether the configured policy expects execution, approval, or block behavior. This reduces the pressure to interrupt a person for every low-risk action.
Show the action and the reason
Present the normalized action, identity, target, relevant dimension scores, policy reference, and evidence limitations. Do not collapse the policy disposition and the reviewer choice into one field.
Record the human decision
Preserve who decided, when, what rationale they supplied, and whether the authorized attempt later executed. A human decision changes the path; it does not erase the original risk signal.
What this page does not establish
- 5D does not supply a universal human-review interface or determine who is authorized to approve.
- A reviewer can make a poor decision, and an approval does not prove the action is correct.
- Human oversight is meaningful only when the reviewer receives enough context and the host preserves the decision.