enterprise-ai-transformation-skills

Use when evaluating an AI idea, AI concept, or early-stage AI proposal — before any pilot design or investment decision. Phrases like "should we pursue this AI idea?", "is this AI concept worth exploring?", "does this AI use case make sense?", "we're thinking about building AI for X — is it a good idea?" all trigger this skill. Diagnoses using a five-role framework — Investigator (real friction), Devil's Advocate (right solution mode), Long-term Strategist (value accumulates), Realist (right capability), Senior Advisor (synthesis verdict). Outputs Fund / Fund-with-condition / Reframe / Kill.

Use when approving or rejecting an AI investment proposal, greenlighting an AI program for funding, evaluating whether an AI initiative should proceed to production, or reviewing an AI business case before board or executive sign-off. Phrases like "should we fund this AI project?", "run the ROI gate on this", "is this AI investment justified?", "we're about to approve AI spending for X", "help me evaluate this AI business case" all trigger this skill. Runs PwC's 20-item Do's-and-Don'ts checklist plus a three-channel ROI model and objective-mode audit. Outputs an approval recommendation with named conditions and red flags.

Use when frontline experts — clinicians, investigators, lawyers, engineers, customer agents — fear AI replacement, deskilling, or accountability shift, and the rollout is stalling on resistance rather than tech. Phrases like "clinicians are afraid AI will replace them", "the team thinks this is automation in disguise", "senior staff are blocking the AI pilot", "we need buy-in from the frontline before we deploy", all trigger this skill. Runs a five-role engagement protocol — empathic listening, jagged-frontier task split, co-design workshop, psychological-safety contract, tradecraft protection — that turns the threatened expert into the co-author of the augmentation. Outputs an Engagement Plan, an AI-leads/Human-leads/Hybrid/Off-limits task split, and a written Safety Contract.

Use when assessing an organization's leadership readiness for AI transformation, surfacing where people and leadership gaps will block AI progress, preparing for a leadership team AI conversation, or diagnosing why AI investments aren't creating value despite good technology. Phrases like "is our leadership team ready for AI?", "surface AI readiness gaps in our leadership team", "help me understand where our people gaps are on AI", "prepare me for a leadership team AI conversation", "why aren't we getting value from AI despite our investment?", "assess AI readiness in our organization" all trigger this skill. Runs a four-role gap analysis (CEO / Manager / Employee / HR) with diagnostic questions for each role. Outputs a 4-role gap report with specific actions per gap.

Use when choosing which AI tool to put in front of a specific user group — a training cohort, a team, a school, a department — or deciding whether to teach deeper prompting, extend existing tools with agent skills, or introduce a new tool. Phrases like "which AI tool should I teach this group?", "choosing an AI tool for our teachers / nurses / analysts", "should I teach advanced prompting or a new tool?" all trigger this skill. Runs six roles — Audience Profiler, Constraint Mapper (hard eliminators before any capability comparison), Intervention-Mode Selector (Deepen / Extend / Introduce), Capability-Delta Assessor, First-Win Realist, Synthesis. Outputs Adopt-now / Adopt-with-scaffolding / Pilot-with-subgroup / Skip plus a growth path. For build/buy/partner sourcing of an AI capability use tech-buy-vs-build; for enterprise-wide curriculum design use people-literacy-curriculum.

Use when designing an AI pilot, structuring a proof-of-concept, or preparing to test an AI idea before full investment. Phrases like "how should we structure this pilot?", "what does a 90-day AI pilot look like?", "help me design the test for this AI initiative?", "we want to pilot AI for X — what do we need?", "what metrics should we track?" all trigger this skill. Runs a structured 7-question pilot design protocol: scope, problem, pre-deployment metrics, blast radius, stop conditions, workflow redesign, ownership. Outputs a one-page pilot brief ready for sponsor approval.

Use when an enterprise has shipped AI agents but cannot see usage, cost, or ROI at the portfolio level. Phrases like "we deployed agents but can't see them", "unclear ROI on our AI", "token spend is exploding", "governance dashboard for AI", "how do we track agent performance", "what's the net ROI after oversight" all trigger this skill. Builds a three-layer KPI tree, dashboard spec, and review cadence that subtracts human-oversight cost from gross gains — outputs a portfolio observability blueprint with sunset criteria and an Observable-and-governed / Partial-observability / Black-box verdict.

Use when deploying an AI agent, reviewing agentic AI governance, checking whether an AI agent deployment is safe and well-governed, or assessing whether an organization's oversight of autonomous AI systems is adequate. Phrases like "check the governance on this agent deployment", "is our agentic AI safe to deploy?", "review guardrails for this AI agent", "we're about to launch an AI agent — what do we need to have in place?", "help me assess our agentic AI oversight" all trigger this skill. Runs IMDA's four-dimension agentic governance check across risk bounding, human accountability, technical controls, and end-user responsibility. Outputs a governance readiness assessment with gaps and required actions.

Use when matching enterprise data sensitivity to AI deployment pattern, deciding whether a use case can run on consumer ChatGPT vs enterprise SaaS vs VPC vs air-gapped, suppressing shadow AI, or specifying the control stack required per data class. Phrases like "can we use ChatGPT for this?", "is this data safe for an enterprise SaaS LLM?", "what controls do we need for PII/PHI in this AI workflow?", "shadow-AI is everywhere, how do we sanction tools?", "what deployment pattern fits regulated data?" all trigger this skill. Maps each data sensitivity class to one of five deployment tiers, names the minimum control stack per tier under NIST AI RMF / ISO 42001 / EU AI Act, and outputs a deployment-pattern verdict (Approved / Conditional / Blocked) with a control checklist and regulatory citations.

Use when plotting an organization's AI maturity stage, assessing where an org sits on the AI capability curve, benchmarking AI progress against industry peers, setting AI strategy ambition level, or preparing a board-level AI readiness briefing. Phrases like "where are we on the AI maturity curve?", "assess our AI maturity", "benchmark our AI progress", "what stage are we at for AI?", "help me understand how mature our AI capabilities are", "prepare a board briefing on our AI readiness" all trigger this skill. Runs MIT CISR's four-stage model alongside Accenture's Foundation × Differentiation 2×2, surfaces the binding constraint (platform gap vs. strategy gap), and outputs a maturity placement with a prioritized roadmap to the next stage.

Use when a leader asks for named peer cases, comparable deployments, or analog stories before making a selection or build decision. Phrases like "how have others done this?", "show me cases from similar orgs", "what peer examples exist for this use case?", "I want to learn from others' experiences", "exchange best practices on AI deployment", "any Stanford / named case studies on this?" all trigger this skill. Retrieves 3-5 named peer cases matched on vertical / org-size / use-case archetype, extracts cross-case patterns against the 95/5 base rate, and outputs a curated case bundle plus a closest-match action recommendation.

Use when a leader cannot decide where to point AI first across a function or business unit. Phrases like "how to focus on areas where AI can be utilized", "finding a useful productive case use", "knowing what to adopt and how to use agentic", "decide among the myriad of choices", "which use case should we pilot first", all trigger this skill. Generates, scores and ranks candidate AI use cases through four sequential roles — Value-Pool Mapper, Capability Archetype Classifier, Feasibility & Moat Scorer, Portfolio Sequencer — and outputs a TOP-3 ranked portfolio with first-pilot pick, why-not-others, and kill list.

Use when designing AI literacy training, addressing workforce that "uses AI as a chatbot only", responding to EU AI Act Art. 4 mandatory literacy duty, building role-based AI upskilling, or fixing low-awareness symptoms across executives/managers/frontline. Phrases like "our employees only use AI as a chatbot", "we need an AI literacy program", "what does EU AI Act Art. 4 require us to train", "build a role-based AI curriculum", "how do we get older workers up to speed on AI", "design an AI training rollout" all trigger this skill. Builds a four-pattern mental-model taxonomy (chatbot / RAG / workflow / agent) crossed with role segments, then outputs a curriculum spec, EU AI Act Art. 4 compliance footprint, and 30-day rollout plan.

Use when moving an AI prototype past demo-ware into production with reliability, fallbacks, and on-call. Phrases like "demo works but production breaks", "need SLOs and evals", "AI still makes mistakes so benefit unclear", "fallback and human-in-loop design", "staged rollout plan", "go-live readiness gate" all trigger this skill. Runs the 5-stage pilot-to-production playbook anchored in Stanford's 51-deployments postmortem and NIST RMF MANAGE controls. Outputs a go-live verdict, remediation list, and week-by-week rollout schedule.

Use when deciding whether to build AI capability in-house, buy it from a vendor, or partner for it. Phrases like "should we build or buy this AI?", "should we train our own model or use an API?", "which vendor should we use for AI?", "help me decide on the AI sourcing strategy", "we're choosing between building vs. buying AI — which is right?" all trigger this skill. Applies the NANDA 2:1 evidence and vendor-archetype taxonomy to produce a BUILD / BUY / PARTNER recommendation with investment rationale.

Use when assessing an organization's AI technology infrastructure, diagnosing why AI pilots aren't scaling, evaluating technical readiness before a major AI investment, or identifying the weakest layer in the AI stack. Phrases like "diagnose our AI tech stack", "why isn't our AI scaling?", "assess our AI infrastructure", "where are we technically on AI?", "what are our biggest tech gaps for AI?", "review our AI architecture" all trigger this skill. Walks the full six-layer stack — data, model, orchestration, tools, oversight, operations — surfaces the weakest link, and outputs a prioritized remediation plan.