Operator Agent Skills

From Productivity to Throughput

Source-grounded pack with Plan AI Leverage for Throughput, Review Queue Health After AI, and Map a Throughput Constraint, plus 1 more.

Get this pack

Grounded in the complete From Productivity to Throughput 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 From Productivity to Throughput 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.

Plan AI Leverage for Throughput

Use this workflow when designing an AI intervention against a known workflow constraint to improve accepted work, decision throughput, or learning throughput while controlling WIP, review displacement, quality, and the next constraint.

Output specification

Return an agent_leverage_plan with constraint, throughput_unit, intervention, bounded_job, output_object, decision_or_learning_loop, flow_controls, quality_gates, review_capacity, measures, predicted_constraint, owner, and decision_rule.

Review Queue Health After AI

Use this workflow when AI has increased drafts, options, tickets, code, analyses, or other work in progress and operators need to diagnose congestion, aging, reviewer overload, batch size, intake quality, and output inflation.

Output specification

Return a queue_health_review with queue, arrival_rate, accepted_exit_rate, wip, aging, segments, review_capacity, quality, rework, output_inflation, root_causes, flow_controls, owner, and follow_up.

Map a Throughput Constraint

Use this workflow when mapping one workflow from trigger to accepted outcome to distinguish local AI productivity from system throughput, locate the actual constraint, and predict where pressure will move after intervention.

Output specification

Return a throughput_constraint_map with workflow, throughput_unit, accepted_finish, flow_steps, queues, wip, review_load, rework, original_constraint, ai_intervention, local_effect, system_hypothesis, predicted_next_constraint, and baseline.

Design AI Throughput Measurement

Use this workflow when measuring or auditing whether AI made one workflow faster and better rather than merely busier, using accepted outcomes, cycle and queue time, WIP, review displacement, quality, decisions, learning, cost, and constraint movement.

Output specification

Return an ai_throughput_packet with workflow, throughput_unit, baseline, intervention, local_productivity, system_throughput, flow, quality_adjustment, review_displacement, decision_throughput, learning_throughput, cost_per_accepted_outcome, segments, new_constraint, and decision.

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.

Choose your reading rhythm.

Start with a weekly briefing, add daily notes, or hear only when a durable essay or research update is ready.

You've successfully subscribed to Antoine Buteau
You've successfully subscribed to Antoine Buteau
Welcome back! You've successfully signed in.