THE DUTYGRAPH FIELD GUIDE
Make sense of
AI governance.
Start with the questions. Understand the landscape. Put the limits in writing.
AI governance starts with the work.
A practical guide to AI governance: responsibilities, risk, access, evidence, and the first questions to ask your team.
Read the guide →FROM POLICY TO A TASKBefore an agent acts, define its job.
A practical AI agent governance checklist covering human ownership, tasks, scopes, approvals, exceptions, and lifecycle changes.
Read the guide →THE GOVERNANCE LANDSCAPEOne landscape. Several different jobs.
Explore a source-linked map of AI governance, agent identity, access governance, and DutyGraph’s work-discovery layer.
Read the guide →A PRACTICAL WORKSHEETPut the agent’s boundaries in writing.
Download an editable agent manifest JSON worksheet for human ownership, duties, tasks, software scopes, approvals, and lifecycle review.
Read the guide →PREPARE THE EVALUATIONFind the gaps before the pilot starts.
An AI governance readiness checklist for a scoped enterprise evaluation: sponsor, workflow, evidence, access, data, and review.
Read the guide →A WORKED EXAMPLEFollow the request all the way to the limit.
A practical agent approval workflow: request, work context, scope review, approval, issuance, and evidence of downstream enforcement.
Read the guide →CHECK THE COMBINATIONOne agent should not quietly become both sides of a check.
Understand separation of duties for AI agents with a supplier-record example, combined-scope review, and evidence for human checkpoints.
Read the guide →Keep exploring
AI governance starts with the work.
A practical guide to AI governance: responsibilities, risk, access, evidence, and the first questions to ask your team.
Read the guide →FROM POLICY TO A TASKBefore an agent acts, define its job.
A practical AI agent governance checklist covering human ownership, tasks, scopes, approvals, exceptions, and lifecycle changes.
Read the guide →THE GOVERNANCE LANDSCAPEOne landscape. Several different jobs.
Explore a source-linked map of AI governance, agent identity, access governance, and DutyGraph’s work-discovery layer.
Read the guide →A PRACTICAL WORKSHEETPut the agent’s boundaries in writing.
Download an editable agent manifest JSON worksheet for human ownership, duties, tasks, software scopes, approvals, and lifecycle review.
Read the guide →PREPARE THE EVALUATIONFind the gaps before the pilot starts.
An AI governance readiness checklist for a scoped enterprise evaluation: sponsor, workflow, evidence, access, data, and review.
Read the guide →A WORKED EXAMPLEFollow the request all the way to the limit.
A practical agent approval workflow: request, work context, scope review, approval, issuance, and evidence of downstream enforcement.
Read the guide →CHECK THE COMBINATIONOne agent should not quietly become both sides of a check.
Understand separation of duties for AI agents with a supplier-record example, combined-scope review, and evidence for human checkpoints.
Read the guide →