Score your agentic commerce readiness before you build.
A five-dimension operator scan for European commerce teams: governance, data, human control, measurement, and execution. It returns a PRISM score out of 100, top-three gaps, and an EU AI Act exposure indication. No data leaves the browser unless you explicitly send the result.
Workflow benchmark
Start from the workflow you want governed.
PRISM pre-fills a benchmark for the selected commerce workflow. Adjust each answer to reflect your current operating reality.
VerticalCPG / field sales
Risk classFINANCIAL
ExposureMedium exposure indication
Ask ATLAS before scoring
Ask what to inspect before turning a score into a diagnostic request.
ATLAS can explain the PRISM path, point to the evidence routes, and keep action requests behind founder-reviewed intake.
READ_ONLYReadiness boundary visible
Public ATLAS can route buyers through PHANTOM, PRISM, Trust, GACO, AVO, Protocol Atlas, and diagnostic intake.
Source corpus
Explicit public manifest
Writes
Disabled
Private routes
Blocked
Public PRISM rubric
Five checks before a governed commerce agent is worth building.
The scan rolls up to the same buyer handoff used by the diagnostic path. When the score is complete, use Send this result to HARNEXAto carry the capability gaps and recommended route into the request.
Governance
Governance maturity
Identity, permissions, audit trail, risk class, and revocation are visible before agents operate.
Data
Commerce data readiness
Product, account, stock, promo, and workflow data can support grounded agent decisions.
Control
Human-agent boundary
The agent knows what it can read, propose, approve, and never do alone.
Measurement
Measurement capability
Cost, latency, efficiency, accuracy, reliability, and workflow ROI can be tracked from day one.
Execution
Organizational readiness
The business can move from proof to governed deployment without committee drag.
Diagnostic evidence map10 public prompts sample the 20-point PRISM diagnostic rubric.
The public scan stays light enough to complete quickly, while each prompt is tied to the evidence HARNEXA validates in the paid diagnostic: ownership, data, action boundaries, CLEAR targets, and adoption path.
Governance2 public prompts5 diagnostic evidence checks sampled
Policy owner, agent register, revocation path.
agent_identity record and tool allowlist.
RiskClass matrix and approval policy.
2 more checks in the paid diagnostic.Data2 public prompts4 diagnostic evidence checks sampled
API/export list, data owner, freshness cadence.
Permission map and personal-data notes.
Policy source, owner, version history.
1 more checks in the paid diagnostic.Control2 public prompts7 diagnostic evidence checks sampled
Owner name, role, KPI responsibility.
Sponsor, budget owner, approval path.
Pilot group, feedback loop, rollout criteria.
4 more checks in the paid diagnostic.Measurement2 public prompts5 diagnostic evidence checks sampled
Current cost, time, quality, or conversion baseline.