Agentic Back Office
Source-grounded pack with Audit an Agentic Back Office, Build a Decision Evidence Packet, and Design a Back-Office Loop, plus 2 more.
Get this packGrounded in the complete Agentic Back Office series.
Open the pack, choose a workflow, and follow its setup and input instructions. You can also open each workflow below.
10 source essays · 5 usable skills · reviewed Jul 2026
Use this pack when
Choose this pack when the work touches Agentic Back Office and you need a bounded workflow with explicit inputs, a defined output, and human review gates.
Choose the job in front of you.
5 source-grounded skills. Load one workflow at a time.
Audit an Agentic Back Office
Use this workflow when assessing whether a back-office workflow is ready to launch, widen, or deepen agent delegation by testing control integrity, event replay, reversibility, exception handling, monitoring, and accountable ownership.
Output specification
Return an agentic_back_office_audit with readiness, replay_results, adverse_tests, control_findings, system_integrity, oversight_health, recovery_test, metrics, decision, required_repairs, owner, and next_review.
Build a Decision Evidence Packet
Use this workflow when preparing a review-ready evidence packet for a finance, procurement, legal, people, compliance, or administrative decision while preserving source provenance, uncertainty, policy boundaries, and specialist judgment.
Output specification
Return a decision_evidence_packet with object, decision, requester, owner, reviewer, facts, evidence_inventory, policy_or_playbook_matches, missing_information, conflicts, risk_flags, recommendation, proposed_action, approval_required, source_citations, and audit_fields.
Design a Back-Office Loop
Use this workflow when converting a recurring finance, procurement, legal, people, compliance, or admin request from a ticket queue into a governed operational loop with explicit objects, states, evidence, owners, and actions.
Output specification
Return a one-page back_office_loop_spec with object, candidate_score, states, transitions, evidence, owners, source_systems, allowed_reads, allowed_writes, approval_gates, exceptions, audit_events, rollback, kill_switch, metrics, and loop_owner.
Map Operational Truth
Use this workflow when an internal workflow reads across fragmented systems or risks creating a shadow source of truth, and operators need explicit field ownership, conflict rules, identity attribution, and authoritative write paths.
Output specification
Return an operational_truth_map with rows for object, field, authoritative_system, owner, source_class, allowed_reads, authoritative_write, freshness, conflict_rule, reconciliation_owner, identity_events, and prerequisite_gaps.
Set Delegation Depth
Use this workflow when deciding how far an agent may prepare, recommend, draft, execute, or self-review a back-office workflow based on reversibility, materiality, evidence confidence, reviewer capacity, and observed reliability.
Output specification
Return a delegation_matrix with workflow_class, action, risk_dimensions, mode, threshold, required_evidence, reviewer, review_reason, log, rollback, promotion_criteria, downgrade_triggers, and review_date.
Keep consequential decisions explicit.
These workflows structure analysis and artifacts; they do not authorize autonomous external, personnel, financial, legal, security, or other high-impact action.