Tools

Human Approval Gate Designer

Describe one action an AI workflow may take. The tool recommends a control, explains the rule and produces a gate specification you can copy without sending us your inputs.

Uli PrantzBuilds and operates all-agents
Published
Key takeaways

Key takeaways

  • The designer scores consequence, reversibility, uncertainty and affected scope, then applies hard stops for critical or irreversible high-consequence actions.
  • Its output is a design prompt, not permission to deploy. Legal duties, internal policy and the actual rights of affected people can require stricter control.
  • The exported contract names the placement, reviewer, evidence packet, outcomes, deadline, escalation and resume rule rather than returning a vague risk label.
  • Everything runs in your browser. No input is submitted, no account is needed and the recommendation changes immediately as you change assumptions.
The Human Approval Gate Designer turns one proposed action into a transparent control recommendation and a copyable seven-part gate specification. It scores consequence, reversibility, decision uncertainty and affected scope, then adds hard stops for critical or irreversible high-consequence actions. The calculation runs in your browser. It sends us no input and grants no authority to deploy the resulting design.

Design one approval gate

Use an effect verb: send, publish, change, grant or pay.

Name the role with authority to refuse, not a shared inbox.

Effect flags

Suggested control · score 7

Deterministic checks plus exception approval

Pause immediately before the exact effect, after the proposal and source evidence are complete.

Evidence packet

  • Exact proposed action, target and normalized payload
  • Original source records and relevant before-state
  • Named policy rule, threshold or reason the request was routed
  • Known uncertainty, plausible consequence and safe alternative
  • Final recipient, destination and externally visible wording

Allowed outcomes

  • Approve the exact proposal
  • Edit and create a new proposal version
  • Reject with a reason
  • Return for more evidence
  • Cancel
  • Expire safely
Escalation: Escalate to the Service owner before 1 business hour; expiry must leave the effect unexecuted.
Show the scoring rationale
  • moderate consequence contributes 2 points.
  • partial reversibility contributes 1 points.
  • bounded judgment contributes 1 points.
  • external person contributes 2 points.
  • External effect adds 1 point.

How the recommendation is calculated

The tool uses a deliberately inspectable heuristic. Consequence adds 0, 2, 4 or 6 points. Reversibility adds 0 for complete rollback, 1 for partial repair and 3 when the effect cannot be recalled. Rule-based, bounded and open judgment add 0, 1 and 2. Affected scope adds 0 through 3. External effect and sensitive data add one each.

Scores up to four suggest a bounded automatic path with post-action sampling. Five through seven suggest deterministic checks plus approval for exceptions. Eight through ten suggest blocking approval. Critical consequence, irreversible high-consequence action or a score of eleven or more produces the strictest result: two independently authorized reviewers or prohibition.

The score selects a starting control. The hard stops prevent a simple average from hiding a severe dimension.
ResultDefault runtime behaviorEvidence needed before relaxing it
Automatic with samplingExecute inside fixed bounds, retain rollback and review a stratified sampleLow sampled miss rate and complete, tested reversal
Exception approvalCode passes the normal path; named ambiguity or threshold pausesStable trigger coverage and useful reviewer interventions
Blocking approvalPause the exact effect until an authorized decision arrivesA safer bounded path proven by production sampling
Two people or prohibitRequire independent authority or do not make the action availableGoverning policy and qualified review, not a lower model score

Why the tool asks about the effect instead of model confidence

NIST's AI RMF connects oversight to context, likelihood, impact and risk tolerance. Google Cloud recommends human involvement for high-stakes decisions, safety-critical work and subjective approval. Both point to the action around the model, not a confidence number in isolation.

A confidence estimate can help route bounded cases when it has been calibrated against similar deployed work. It cannot decide whether a public message, access grant or decision about a person is authorized. That is why the designer asks what happens if the action is wrong, who is affected and whether the effect can be reversed.

Use the copied specification as a review prompt

  1. 01Challenge the classification
    Ask the process owner, control owner and affected operator whether consequence and reversibility include downstream use, disclosure and delay.
  2. 02Replace generic evidence with real fields
    Name the source record, before-state, policy clause, recipient, diff and safe alternative the reviewer will actually see.
  3. 03Verify authority and expiry
    Confirm the named role can refuse, define coverage when they are absent and make expiry land in a safe owned state.
  4. 04Implement the state machine
    Use the approval workflow patterns to bind the proposal and resume exactly once.
  5. 05Test and measure
    Try altered, expired, unauthorized and duplicate decisions. Then track coverage, interventions, expiry and sampled misses rather than approval volume alone.

Example interpretation

Customer-facing status update, using the defaultsSimulated example data

The default action has moderate consequence, can only be partly reversed, needs bounded judgment, affects an external person and sends externally. It scores seven, so the tool suggests deterministic checks plus exception approval.

That result is useful only after the team defines the normal path. A templated acknowledgement with no promise may pass automatically. A message that names a cause, compensation or recovery time becomes the exception and waits for the service owner. If every status update makes a public commitment, select the wider affected scope and use blocking approval instead.

Limitations and when not to use this

  • The score is an educational design heuristic, not a risk assessment, legal classification, security review or authorization decision.
  • It does not know your laws, contractual duties, internal policies, reviewer competence, model calibration or actual failure history. Any of those can require a stricter control or prohibit the action.
  • Browser inputs are not submitted to all-agents, but avoid entering confidential records into any planning tool. Describe the action and role, not a real customer, employee or credential.

Sources

  1. AI Risk Management Framework CoreNIST Accessed 30 July 2026
  2. Choose a design pattern for your agentic AI systemGoogle Cloud Architecture Center Accessed 30 July 2026
  3. Regulation (EU) 2024/1689, Article 14EUR-Lex Accessed 30 July 2026

Implement the runtime states

Take the copied specification and map it to durable pending, approve, edit, reject, expire and resume transitions.

Implement the runtime states
About the author

Uli Prantz

Builds and operates all-agents

Uli Prantz builds all-agents, the process-automation platform this site documents. He writes about the operational side of automating recurring business work: where deterministic code beats model judgment, where it does not, and where a human still has to approve.

Bring one process. We will scope it in 30 minutes.

You leave the call knowing whether it is a fit, what can become code and what still needs a person.

Book a discovery call