AI AI Crafters
Agent Risk Assessment Preview

Assess AI agent risk before production.

A static, browser-only mockup of the assessment template. Adjust autonomy, tools, data, impact, and red flags to see the production risk tier and required controls update instantly.

Assessment input

0-3 scoring model

Autonomy

2
ScoreAutonomy levelDescription
0Answer onlyThe agent answers questions and does not take action.
1RecommendThe agent suggests actions, but a human performs them.
2Act with approvalThe agent can prepare or execute actions after explicit approval.
3Act autonomouslyThe agent can act from triggers or complete steps without case-by-case approval.

Tool access

3
ScoreTool access levelDescription
0No toolsThe agent only generates responses.
1Read-only toolsThe agent can retrieve data but cannot change systems.
2Limited write toolsThe agent can create drafts, tickets, notes, or low-impact records.
3High-impact toolsThe agent can change customer, finance, HR, legal, security, or infrastructure state.

Data sensitivity

2
ScoreData levelDescription
0PublicPublic information only.
1InternalNon-public business information with low sensitivity.
2Confidential or personalCustomer, employee, commercial, contractual, or operational data.
3Restricted or regulatedSpecial category personal data, credentials, secrets, payment data, legal privilege, regulated records, or security-sensitive information.

Business impact

2
ScoreImpact levelDescription
0MinimalMinor inconvenience or easily corrected output.
1LowInternal rework, limited user confusion, or small operational cost.
2ModerateCustomer impact, compliance evidence gap, financial loss, data exposure, or process disruption.
3SevereMaterial legal, financial, safety, rights, security, regulatory, or reputational harm.

Critical red flags