Automation
Source-grounded pack with Specify an Automation Workflow, Scan an Automation Opportunity, and Build an Automation ROI Brief, plus 1 more.
Get this packGrounded in the complete Automation series.
Open the pack, choose a workflow, and follow its setup and input instructions. You can also open each workflow below.
10 source essays · 4 usable skills · reviewed Jul 2026
Use this pack when
Choose this pack when the work touches Automation and you need a bounded workflow with explicit inputs, a defined output, and human review gates.
Choose the job in front of you.
4 source-grounded skills. Load one workflow at a time.
Specify an Automation Workflow
Use this workflow when specifying a production automation or bounded agent workflow with contracts, durable state, idempotent side effects, retries, queues, permissions, stop conditions, observability, replay, evaluation, security, and launch controls.
Output specification
Return an automation_workflow_spec with scope, boundary, contracts, states, idempotency, retries, queues, model_jobs, agent_permissions, review, observability, security, failure_controls, evaluation, launch_mode, owner, and retirement_trigger.
Scan an Automation Opportunity
Use this workflow when decomposing a workflow into deterministic, probabilistic, and accountable steps so operators can decide what code, a model, or a human should own before choosing tools or building automation.
Output specification
Return an automation_boundary_map with workflow, trigger, final_state, steps, work_type, owner, reason, inputs, outputs, risk, reversibility, controls, prefilters, exceptions, recommendation, and prerequisites.
Build an Automation ROI Brief
Use this workflow when deciding whether an automation workflow is economically and operationally worth building by comparing baseline labor and failure costs with deterministic coverage, model cost and latency, review load, maintenance, risk exposure, and staged benefits.
Output specification
Return an automation_roi_brief with baseline, architecture_mix, cost_model, benefit_model, scenarios, risk_adjustments, capacity, latency, proof_stage, stop_condition, decision, and sensitivity_drivers.
Design Human Review in Automation
Use this workflow when designing selective human review for an automation workflow, including trigger reasons, evidence-rich queue items, decisions, service levels, correction capture, sampling, escalation, and feedback into rules and evaluations.
Output specification
Return a human_review_design with triggers, decision_options, queue_schema, reviewers, sla, backlog_controls, escalation, correction_taxonomy, sampling, metrics, and feedback_loop.
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.