Stage 01 of 05 · The maturity model

Stage 01 Experimental

Isolated agent use by individuals.

  1. 01 Experimental
  2. 02 Shared
  3. 03 Operational
  4. 04 Governed
  5. 05 Production-Ready
Stage 01

Experimental

Isolated agent use by individuals.

A handful of engineers and analysts run agents from personal accounts to save themselves time. Nothing is registered, reviewed, or shared. Ask "which agents are running here?" and nobody can answer.

What makes it different

Nothing is written down. Capability lives in individual heads and personal API keys.

Primary risk

No inventory, personal tokens, no review. Agents are shadow infrastructure from day one.

Next control to fix

Start an inventory — even a spreadsheet. You cannot govern what you cannot list.

What the controls look like at this stage

Maturity is measured across six control surfaces. Here is where a immature organisation — one sitting at Stage 01 — typically stands on each.

Inventory

Immature

No list. Nobody can say which agents exist, what they touch, or where they run.

Ownership

Immature

Personal accounts. No named human is accountable for any agent’s behaviour.

Runtime visibility

Immature

None. What an agent does between input and output is invisible.

Access control

Immature

Personal tokens with broad scopes. Agent identity is indistinguishable from a user’s.

Evaluations

Immature

None, or a single check before launch that is never repeated.

Governance

Immature

No policy, no audit path, no way to stop an agent cleanly.

Advancing to Stage 02 — Shared

You do not skip stages. You close the gap in front of you, and only one control matters most right now.

Do this next

Start an inventory — even a spreadsheet. You cannot govern what you cannot list.

02 Shared Agents reused across teams. See Stage 02 →

Stage 01 — Experimental, answered.

Short answers for teams placing themselves at the Experimental stage of the agent operational maturity curve.

What is Stage 01 (Experimental) of the agent maturity model?

A handful of engineers and analysts run agents from personal accounts to save themselves time. Nothing is registered, reviewed, or shared. Ask "which agents are running here?" and nobody can answer. In short: isolated agent use by individuals.

What is the main risk at the Experimental stage?

No inventory, personal tokens, no review. Agents are shadow infrastructure from day one.

How do you move beyond the Experimental stage?

Start an inventory — even a spreadsheet. You cannot govern what you cannot list. That is what takes an organisation from Stage 01 (Experimental) to Stage 02 (Shared).

Is your organisation at Stage 01?

Take the free Agent Operational Maturity Assessment — seven minutes, no email required to see your stage and the highest-risk gap to close next.

Take the assessment →