Magnus HedemarkandGitHub
94b7231147
* feat: add AI operating economics skill Add an evidence-led methodology for evaluating AI workflow value, cost, worker effects, quality guardrails, and authority expansion. Includes research references, durable decision templates, and six eval cases. AI assistance: Jasper, on behalf of Magnus Hedemark. Signed-off-by: Magnus Hedemark <magnus919@pm.me> * fix: resolve AI economics review findings Align section numbering, evidence-language examples, and intervention-mode terminology identified by the exact-head review. Signed-off-by: Magnus Hedemark <magnus919@pm.me> --------- Signed-off-by: Magnus Hedemark <magnus919@pm.me>
3.0 KiB
AI Operating Economics
A decision method for determining whether an AI-enabled workflow is creating value at an acceptable cost, quality, and human-impact boundary.
Why Install This Skill
AI pilots often produce an attractive number: faster handling time, more tasks completed, lower apparent cost, or high adoption. That number is rarely enough to decide whether the intervention should scale. It may omit review work, infrastructure, quality loss, worker differences, customer effects, or the cost of changing the surrounding process.
This skill helps an agent connect those dimensions into one accountable decision. It distinguishes measured workflow evidence from vendor claims, separates speed from value, inspects who benefits and who bears the cost, and produces a bounded recommendation: scale, constrain, redesign, hold, retire, or exception.
After installing it, your agent can prepare an AI initiative evidence record, challenge weak ROI claims, design a value-realization review, and tell you exactly what evidence is missing before more authority or spend is granted.
What You Get
| File | What it provides |
|---|---|
SKILL.md |
Core routing, operating principles, nine-step workflow, evidence classes, dispositions, pitfalls, and verification checklist |
references/evidence-method.md |
Detailed comparison design, cost-boundary, worker-impact, uncertainty, and learning-loop method |
references/source-index.md |
Primary and independent sources with claim scope, caveats, and permitted use |
templates/ai-initiative-evidence-record.md |
Fillable record for one AI workflow or use case |
templates/ai-economics-review.md |
Executive or lifecycle review template for one or more initiatives |
evals/evals.json |
Six output-quality cases covering ROI claims, heterogeneous effects, incomplete TCO, vendor evidence, authority, and retirement |
Quick Start
No setup, API keys, or runtime dependencies are required.
Ask your agent:
Review this AI pilot and tell me whether we should scale it. Separate measured outcomes, cost, quality countermetrics, worker effects, evidence gaps, and the authority we should grant next.
For a durable review, ask it to use templates/ai-initiative-evidence-record.md and save the completed record in your normal project documentation system.
Triggers
Load this skill when you need to:
- Evaluate an AI use case, pilot, agent, or automation for value realization
- Decide whether to scale, constrain, redesign, hold, or retire an AI workflow
- Review AI productivity, savings, adoption, or transformation claims
- Connect AI cost attribution to workflow outcomes and quality
- Assess worker, customer, user, or distributional effects of an AI intervention
- Prepare an AI business case or post-launch value review
Requirements
- No runtime dependencies or external services
- A stated workflow and decision owner are strongly recommended
- Financial, statistical, analytics, evaluation, governance, and runtime details are routed to adjacent skills