Generated, not gathered
The trail is written by the system as it runs. Nobody assembles it, nobody remembers to turn it on, and nobody can choose not to.
Fails when: a script exports logs the week before an audit.
Euler Center · Work Composition · The gate
Is this workflow ready for an AI agent?
Use six checks to examine who can decide, intervene and stop the work, and what evidence supports deployment. Under this method, all six cells must be solid before proceeding. Your accountable team owns the deployment decision and any additional requirements.
Run itPer deployment, not per program
ItemsSix. All must be solid.
PurposeA readiness review for one deployment
CostFree, no signup. Forward it to whoever needs it.
I / The six items
Check responsibilities, controls and evidence
01Authority is written, not assumedA named list of what this agent may decide alone, what it must route to a person, and who can halt it mid-run. If the halt is theoretical because the process finishes in eight seconds, the cell is blank. Required by the human oversight duty.
02Composition matched to problem typeEvery task the agent touches has been placed by problem type. Nothing agent-led sits in a complex or chaotic domain.
03The escalation trigger fires by itselfA machine-detectable condition surfaces the exception before a person notices. An andon cord nobody can see is decoration.
04Evidence is produced unpromptedTrail, reasoning log, and input lineage generated as a byproduct of running. Reconstructed evidence is not evidence.
05A named person is qualified and availableNot a role in a policy. A person, assessed on this workflow against something like a defensible judgment taxonomy, rostered while the agent runs, and empowered to override without asking first.
06Rollback is rehearsedSomebody has actually stopped this agent outside production and restored the prior state. A documented rollback never executed is a hypothesis.
Under this method, a blank or thin cell means pause and address the gap.Map it to your regime
Use the checklist to surface gaps and record decisions. Passing it is not certification, legal approval or a guarantee of safe operation. The framework references below provide research context for a review; they do not establish which obligations apply to your deployment.
II / Framework mapping
Use framework references to guide the review
These research mappings point to questions about oversight, records and controls. The frameworks differ in scope and purpose. Confirm applicability and current requirements with the people responsible for compliance in your organization.
| Regime | What it asks for | Cells it lands on | Evidence to examine |
|---|---|---|---|
| EU AI ActArticle 14 and Annex III | A person who can understand the system, intervene in it, and stop it. Employment uses are classed high-risk. | Authority, Mastery, Evidence | Named oversight role, intervention log, halt mechanism that is actually reachable in the runtime |
| ISO/IEC 42001AI management system, 2023 | A certifiable management system: policy, roles, risk treatment, and continual improvement around AI. | Evidence, Purpose, Learning | Documented management system, competence records, internal audit trail |
| NIST AI RMFVersion 1.0, 2023 | Practice organized around govern, map, measure, and manage. Voluntary, and widely used as the common vocabulary. | All twelve, weighted to Evidence | Risk mapping per use case, measurement plan, documented management decisions |
| CMMCFor the US defense base | Assessed cybersecurity practices, with certification levels and third-party assessment. | Evidence, Authority, Resources | Control implementation evidence, scoped boundary, assessed rather than self-declared |
This table is a research aid, not a complete compliance checklist. Check the primary sources for current requirements and dates. The library entry provides the supporting references and limitations.
III / What evidence actually means
Produced as a byproduct, not assembled before the audit
Plan how the workflow will produce records of decisions, interventions and outcomes. These four practices help a team examine what happened and judge whether the work improved.
Reasoning, not just outcome
What the agent had in front of it, what it weighed, and what it rejected. An outcome log tells you what happened and nothing about whether it should have.
Fails when: the record is the decision without the inputs that produced it.
Lineage on the inputs
Which data, which version, which model, which prompt or policy was in force. Most disputes are about the inputs rather than the logic.
Fails when: the model was silently updated and nothing recorded the change.
A named person against each loop
Not a role in a policy document. A person, assessed for this workflow, rostered while the agent runs. This is what stops blame landing on whoever was nearest.
Fails when: the accountable party is a team, a committee, or a job title.
IV / Who is behind this
Work, people and evidence
Work redesign
Euler Center helps teams examine one workflow, clarify responsibilities and decide what should change. The Work Composition Canvas is one tool for that work.
Team preparation
Reviews, workshops, design sessions and tabletop exercises help people practice the decisions and exceptions their work requires. Methods are agreed within the engagement scope.
Measurement
Define what evidence would show improvement, examine results and decide what to expand, change or stop. A favorable finding is not promised.
Experience
Founder Sravan Ankaraju brings experience from Microsoft Enterprise Services, Lockheed Martin and workforce transformation at Divergence Academy, alongside current AI engineering and teaching.
V / Run the gate with us
Bring the decision you need to make
Use the gate yourself, or ask Euler Center about a scoped review of your workflow. We agree the appropriate engagement, evidence requirements and fees before work begins. Share a brief description to start the conversation.
See the existing engagement model and pricing.
Govern
Decide what to expand, change or stop.
The Work Performance Review examines one workflow and the evidence needed for a decision. It is a scoped engagement, not an individual credential.
The Work Performance ReviewLearning and practice