Compliance & Risk

Control Attestation Agent

Finds the controls that exist on paper and nowhere else

Version 2Published Sep 2, 2026
Watch it work▶ Try it on your own figures — free

6 of 9 controls failed their last test or are overdue for attestation.

9 controls checked. 6 are flagged for 14 points; the longest gap is 59 days.

Start with IR-01 Incident drill. The owner with the most flagged controls is Dana.

A passing test does not prove a control is well designed: that needs an assessor.

14
6/9
Controls flagged
6
9
Risk points
14
Longest (days)
59
Share flagged
67%

Why rows were flagged

6
  • Attestation overdue67%
  • Last test failed33%

Value flagged by Owner

  • Dana2
  • Yusuf1
  • Sara1
  • Omar1
  • Lina1

Items to act on, worst first

ItemOwnerValueDays past attestationWhy
IR-01 Incident drillYusuf359Last test failed
AC-02 Privileged accessDana320Last test failed
HR-01 Joiner/leaver checksSara250Attestation overdue
CM-02 Emergency changesOmar241Attestation overdue
AC-01 User access reviewDana231Attestation overdue
BC-01 Backup testLina231Attestation overdue

What to do this week

  • Fix and retest IR-01 Incident drill first: its last test failed.
  • Ask Dana to attest every overdue control this week.
  • Schedule attestations on the calendar with the control owner, not from memory.

Do

  • Fix the highest-risk item first, then run the check again on fresh data.

Don't

  • Don't act on one row alone: open the source record first.

How to improve

  • Add more detail per row (owner, date, source) so the next report can trace each item.
  • Add a second period of data so the next report can show whether things are improving.

Demo

Try it now — free

Type your figures (or fill in samples) and get the answer in a minute.

Start from a template

Make it decision-ready

Also include

No card · no API key · your figures are not stored

Overview

Compliance programmes accumulate controls. Each was created for a real reason, written into a policy, assigned an owner and a frequency, and then left. Nobody removes one, because removing a control means explaining why it is unnecessary — harder than leaving it. The register grows every year and the capacity to perform them does not.

What follows is predictable. Controls owned by people who left. Controls performed but never evidenced, which is indistinguishable from not performed when an auditor asks. Quarterly controls performed twice a year. Controls requiring an independent reviewer where the reviewer is the person who did the work — the segregation failure that turns a control into paperwork. And attestations signed in bulk on a deadline, which is a signature rather than a check.

The dangerous version is the documented control nobody performs, because the organisation believes a risk is covered when it is not. An undocumented gap is at least known.

This agent reads the register against what was actually done and reports which risks are covered only on paper. It never asserts that a control was effective — that judgement requires evidence and accountability belonging to a person who signs their name to it.

Benefits

You know which risks are actually covered

Rather than which ones have a policy about them.

Findings before the auditor's

And a register you can defend.

A case for removing controls

The ones nobody performs and nobody misses.

What it does

  • Finds controls that exist only on paper

    The dangerous kind. The organisation believes a risk is covered when it is not, and an undocumented gap is at least known.

  • Separates performed from evidenced

    A control performed and never evidenced is indistinguishable from one never performed, when asked.

  • Catches self-review

    An independent check performed by the person who did the work. The control exists and does nothing.

  • Detects bulk attestation

    Forty sign-offs in one minute on a deadline is a signature, not a check.

  • Finds orphaned controls

    Owned by someone who left. Nobody notices, because a control with no owner produces no complaints.

  • Asserts no effectiveness

    Whether a control worked is a judgement someone signs their name to. It reports what was done.

What this changes

What this changes about your week

It takes over
Checking whether documented controls are actually performed and evidenced.
Done today by
Attestation, which is a signature rather than a check.
How often
Quarterly, on a deadline.
Typical manual effort
Every control cross-checked against performance records, owners and independence.

Why Compliance & Risk in particular. A documented control nobody performs is worse than no control, because the organisation believes a risk is covered. An undocumented gap is at least known.

  • Runs unattended

    Started by you or by an event, and it finishes on its own. Nothing waits for someone to be at a desk.

  • The same standard every time

    The two-hundredth item is held to the bar the first one was. Consistency is the part people cannot sustain.

  • It cannot act on its own

    Control Attestation Agent has no path to sending, spending or committing. That limit is why its output is safe to act on.

How it runs

This agent runs server-side through the PROMIVO runtime. Each run is logged step by step and every tool call is permission-checked before it executes.

  1. 1InputValidated against the agent input schema
  2. 2Modelclaude-opus-5 with automatic fallback
  3. 3Toolssearch_knowledge, extract_structured_data
  4. 4PermissionsRuns READ actions automatically; anything higher waits for approval
  5. 5OutputStructured result matching the output schema
  6. 6LoggingTokens, duration and every step recorded

Requirements

Integrations

No integrations required.

Before you start

  • The control register: control, owner, frequency, independence requirement
  • Performance records with dates and who performed them
  • Attestation records with timestamps
  • Current staff list
  • Requires the All-Access plan or higher

Use cases

Quarterly control review

What was performed, evidenced and by whom.

Before an audit

The gaps a reviewer would open with.

After a reorganisation

Which controls lost their owner in the move.

Rationalising a register

Which controls earn their place.

Pricing

All-Access

Billed monthly through your PROMIVO subscription. Cancel at any time.

Runs consume your plan allowance for agent executions and tokens. See plan limits

FAQ

Does it decide whether controls are effective?

No. Effectiveness is a judgement with accountability attached to whoever signs it. This reports whether a control was performed, evidenced, and by someone independent — the facts that judgement needs.

Does it perform or sign off controls?

Never. It signs nothing, attests to nothing and closes nothing. A control signed by an agent is not a control.

Does it judge individuals?

No. It reports control-level findings. Where a pattern involves one owner it reports it as a capacity or design question, because a person owning forty controls is a register problem, not a personal one.

Reviews

Reviews open once customers have run this agent.

Need this agent to work differently?

Tell us what to change and our team will scope a customised version for your business.

Customize this agent