How much does ApexClaw cost?

There is no public price card because scope varies by agent count, action classes governed, and integration depth. Pricing is set after the Agent Trust Gap Brief establishes the actual surface. What is fixed is the shape: platform, integration, and operation.

Why no fixed price online?

A number without a scope is a guess. The Gap Brief takes two business days and requires no system access, and it produces the inventory that scoping depends on. Pricing follows evidence, not the reverse.

What drives the cost?

Three layers: platform licensing (scales with agents and action volume), integration engineering (placing gates in the effect path - usually the largest line), and operation (policy authorship, approvals, evidence review).

Is there a minimum useful engagement?

Yes: governing one irreversible action class end to end - typically sending or paying - with deny-by-default policy and signed receipts. It bounds cost while producing defensible evidence immediately.

Do you offer the Gap Brief for free?

The Agent Trust Gap Brief is the low-friction front door: a fast read of your publicly and structurally observable agent risk surface, with clear limitations stated. It is where scoping and pricing begin.

Home › Pricing

Pricing

How ApexClaw is scoped and priced

Pricing follows evidence. There is no public price card because a number without a scope is a guess - so scope comes first, through a two-business-day brief that needs no system access.

Get an Agent Trust Gap Brief

The three layers every quote is built from

Platform

Licensing for the control plane: identity, policy evaluation, receipts, approvals. Scales with agent count and action volume. Usually the smallest of the three.

Integration

Engineering to place enforcement in the effect path - the component that holds the credential - rather than in agent instructions. The dominant cost in most deployments, and the reason starting narrow wins.

Operation

Policy authorship, approval-queue staffing, evidence review, and the quarterly proof that controls still fire. A control with zero recorded refusals is unproven; producing that evidence is an operating activity.

Why scope precedes price

The variable that moves cost most is the action inventory: what each agent can cause that cannot be undone. Most organisations have never written that list, and it determines every control that follows. The Agent Trust Gap Brief produces it in two business days with zero access required - which is why pricing starts there rather than with a number pulled from the air.

The bounded starting point

Govern one irreversible action class end to end - most teams pick outbound sending or payment - with deny-by-default policy, payload-bound approval, and a signed receipt per execution. It caps all three layers while proving the model and producing evidence from day one. Deeper coverage then proceeds action class by action class.

Last verified 2026-08-12. Structural rather than currency-specific: integration and operating costs vary with stack and sector. For scoped numbers, start with the Gap Brief.

The three layers, compared

Layer	What it covers	Typical share of cost
Platform	Identity, policy evaluation, receipts, approvals	Usually smallest
Integration	Enforcement placed in the effect path, not agent instructions	Usually dominant
Operation	Policy authorship, approval-queue staffing, evidence review, quarterly proof	Ongoing

Common questions

Why is there no public price card?

A number without a scope is a guess, and guessed prices get anchored on and argued about later. The Gap Brief establishes real scope in two business days before any figure is discussed.

How long does scoping take?

Two business days for the Agent Trust Gap Brief, which needs no system access. Deeper scoping for a full audit or a governed pilot follows from what that brief finds.

What is the bounded starting point?

One irreversible action class, governed end to end with deny-by-default policy, payload-bound approval, and a signed receipt per execution. It caps cost while proving the model works.

Does pricing vary by region?

Not structurally. Integration and operating costs vary with stack and sector rather than geography, though local obligations like the EU AI Act shape which controls are mandatory versus optional.

Sources

The cost of skipping scope is not abstract: the EU AI Act's three penalty tiers run up to 7%, 3% and 1% of global turnover depending on the obligation breached, and Gartner projects more than 40% of agentic AI projects will be cancelled by the end of 2027, disproportionately the ones that never wrote down an action inventory. Machine identities already outnumber human ones by more than 80x in most enterprises, which is most of why "scope it later" is the expensive option. Standards referenced: OWASP Top 10 for Agentic Applications, NIST AI RMF, and ISO/IEC 42001 for organisations that need a management-system reference point rather than a control-by-control one.

Why this shape, not a tiered price list

A tiered subscription price list assumes every buyer's agent estate looks roughly the same, which is exactly the assumption the Agent Trust Gap Brief exists to test before anyone commits to a number. Two organisations of identical size can have wildly different pricing outcomes depending on how many irreversible action classes their agents actually touch — payments, outbound communication, infrastructure changes, data deletion — and how much of that already sits behind a real effect-boundary control versus behind agent instructions alone, which is not a control at all.

This is also why the platform, integration and operation layers above are kept separate rather than bundled into one number. Platform licensing is the part every vendor can quote instantly because it scales predictably with agent count. Integration is the part that varies most, because it depends on how deeply enforcement has to reach into an existing stack to sit in the actual effect path rather than in a prompt. Operation is recurring by design: a control that has never produced a refusal receipt has never been tested, and producing that evidence on a cadence is an ongoing activity, not a one-time deliverable.

The practical effect is that the fastest path to an accurate number is the same two-business-day brief that requires no system access in the first place — not a longer sales cycle, and not a published price list that would have to average away the real variance between a five-agent pilot and a five-hundred-agent estate.