Skip to content
Start

AI Readiness Assessment

2 to 30 days

Three lengths, from two days to thirty. Each ends with where you stand on the five questions and what to fix first.

Who this is for
  • Leadership told you to "figure out AI," and you need a plan you can defend.
  • You think AI agents are spreading through departments, and nobody's tracking what data they touch.
  • You have one pilot that's stalled in security review and you don't know why.
  • Something's forcing the question: an audit finding, a board question, a customer security questionnaire, or an incident.
And who it isn't
  • You're already past discovery and need to ship a specific AI use case. Start with the Sprint.
What we do

Which length fits?

2 days: Agentic Trust Diagnostic
Pick it when
You think you have a problem and need to know fast.
What we cover
One business unit, or a set of agents you name.
What we do
Two days of working sessions with your security and AI leaders.
What you walk away with
A one-page heatmap (a color chart of where your risks are) and a 30/60/90-day roadmap. Both arrive within five business days.
See the 2-day page
7 days: Agentic Trust Review
Pick it when
You know you have a problem and need it in writing for leadership or a regulator.
What we cover
All five questions, scored.
What we do
Interviews across five teams and a review of your documents and controls. Plus a count of your agents.
What you walk away with
A written report. Every finding has a risk rating, and they're ranked in order.
Jump to the Review
30 days: Agentic Trust Blueprint
Pick it when
You need the full picture and a plan with owners.
What we cover
All 25 requirements in the Agentic Trust Framework.
What we do
Everything in the Review, then a check against each requirement, one by one.
What you walk away with
A phased plan with an owner on every step, plus the interactive final report and a short executive deck.
Jump to the Blueprint

The five questions map to the Agentic Trust Framework's five elements: Identity Management, Behavioral Monitoring, Data Governance, Segmentation, and Incident Response.

What does the 7-day Agentic Trust Review add?

The Review is for when you already know there's a problem and need it written down. It includes everything in the Diagnostic. Then we interview people across security, AI, identity, compliance, and legal, and we review your documents and controls. Your team counts your agents using our method. We write down what each agent may decide on its own and where a person's approval has to interrupt it. You get a score on each of the five questions and a written report. Every finding has a risk rating, and they're ranked so you know what to fix first. The report is written to what an auditor or examiner asks: can you track each agent, can you limit what it does, does a person approve the key decisions, can you stop it, and can you prove all of it. It still holds up if you change vendors next year.

How does the agent count work?

Your team pulls the data, using our method. Agents leave tracks in six places: your identity provider (your login system), your secret store (the vault for software passwords), your build pipeline accounts, your AI gateway logs, your laptop and server data, and your expense reports. We sort what comes back into four findings. Agents logging in as a person. Agents sharing one login. Keys that never expire. Agents whose owner already left.

See how the agent count works

What does the 30-day Agentic Trust Blueprint add?

The Blueprint is the full picture, plus a plan in the right order. It includes everything in the Review. Then we check you against all 25 requirements in the Agentic Trust Framework, one by one, and show where you are today next to where you need to be. We write the operating model your teams will run: one owner per agent, an expiry date on every agent, how much an agent may do on its own at each level, and how a new one gets approved. We set the license strategy, so you know which agent-security add-ons from Microsoft, Okta, Google, or AWS you need and which you can skip. You walk away with a phased plan that has a named owner on every step, a technical appendix for your engineers, the interactive final report, and a short executive deck.

What's the interactive final report?

It's a single file that opens in a browser. It holds everything from the engagement: the meetings, the decisions, the outcomes, the deliverables, and the findings. Your leaders can read the top. Your engineers can click down into the detail. It comes with the Blueprint and with every larger engagement.

What does leadership get to track?

Four numbers, reported every quarter once the plan starts. Agents with a named owner. Agents with an expiry date. High-impact agents with a person approving their actions. The difference between what discovery finds and what's registered. The last one shows you the agents that got in without approval.

What your team learns

How to keep the scorecard current after we leave. How to explain where your agents stand to your executives without using the word "governance."

Duration
2 to 30 days
Your team
Two days of working sessions for the 2-day. A few hours a week from leadership and one or two technical leads for the longer two.
Proof

One we've done before.

Years stalled. Unblocked.

An aerospace defense contractor had critical work stalled for years, knotted up on identity. We got a heavy Zero Trust platform running inside a brutal compliance environment, then designed the strategy that had been missing. Zero Trust means never trust, always verify. We still build and lead their identity architecture today.

See all four stories →

Questions we get

Real objections, answered straight.

  • How is this different from a NIST AI RMF audit?
    NIST tells you what to think about. We tell you where you stand and what to do Monday. We use NIST, ISO 42001, and OWASP as inputs; the deliverable is your roadmap, not a compliance report.
  • Will you push us toward your own product?
    No, because we don't have one. We don't resell, take referrals, or partner with vendors. If the right answer is "your existing stack does this," we say so.
  • What if we have nothing in place yet?
    That's fine. The Assessment works for "we have one pilot stuck in security" and "the board asked about AI last week" equally well. We start where you actually are.
  • Does the $5,000 Advanced ATF Assessment count toward this?
    Yes. The full $5,000 counts toward any engagement you sign within 30 days of getting your report. It applies once.
  • What does it cost?
    We quote each length after we scope it with you. It's one set price for a defined piece of work. The only price we publish is the $5,000 Advanced ATF Assessment.

Want to know if this is the right size?

One 20-minute call. We'll tell you straight. If a different engagement fits better, or if we're not the right people at all, we'll say so.