Prompt Kit
The Age of Intent Engineering Prompt Kit
Prompt Kit: The Age of Intent Engineering
This kit turns the intent engineering framework into working tools. It helps you diagnose where your AI deployments (or personal AI use) are optimizing for the wrong objectives, build machine-readable intent layers that encode what you actually want AI to optimize for, and map your workflows against the three-layer architecture the article describes. Five prompts span both individual and organizational scale — starting with a 10-minute quick version for anyone short on time.
How to use this kit
Short on time? Start with Prompt 1. It's designed to run in 10 minutes and will tell you where your biggest intent gap is — personally or organizationally — with concrete next steps you can act on today.
Individual contributors: Prompts 1 and 2 are for you. Prompt 2 builds a personal intent layer — a structured document you can paste into any AI session so the AI understands your goals, preferences, and decision boundaries persistently instead of starting from zero every conversation.
Leaders deploying AI at scale: Prompts 3, 4, and 5 are your build sequence. Prompt 3 diagnoses your organizational intent gap across all three layers. Prompt 4 generates agent-actionable intent specifications for specific deployments. Prompt 5 maps your workflows into agent-ready, agent-augmented, and human-only categories with intent requirements for each.
Recommended tools: These prompts work best in thinking-capable models like ChatGPT, Claude, or Gemini, since they require the AI to hold complex organizational context and reason through tradeoffs. Use whichever you prefer — the prompts are model-agnostic.
Prompt 1: 10-Minute Intent Gap Diagnostic
Job: Rapid diagnostic that identifies where your biggest AI intent gap is — individually or organizationally — and gives you a prioritized action plan you can start this week.
When to use: You've read the article and want to know where you stand without a multi-hour exercise. Or you're trying to quickly frame the problem for your team or leadership.
What you'll get: A scored assessment across the three intent layers (context infrastructure, workflow coherence, intent alignment), your single highest-risk gap, and 1-3 specific actions ranked by impact and effort.
What the AI will ask you: Whether you're diagnosing individual or organizational AI use, what AI tools/agents you're currently using, what you're trying to accomplish with them, and where things feel off.
Prompt 2: Personal Intent Layer Builder
Job: Creates a structured, reusable "intent document" — a personal operating manual for AI collaboration that you can paste into any AI session so the AI understands your goals, priorities, decision style, and boundaries without you re-explaining them every time.
When to use: You're tired of starting every AI conversation from zero. You want AI to operate as an aligned collaborator, not a capable stranger. You want to move from reactive prompting to proactive, intent-aligned AI use.
What you'll get: A structured personal intent document covering your role, goals, priorities, decision preferences, communication style, and autonomy boundaries — ready to paste into any AI conversation as persistent context.
What the AI will ask you: Your role, what you're trying to accomplish this quarter, how you prefer to make decisions, what you want AI to handle independently vs. flag for you, and what "good work" looks like in your world.
Prompt 3: Organizational Intent Gap Audit
Job: Assesses your organization's AI deployments against the three-layer intent engineering identifies where you're most vulnerable to the Klarna problem — AI succeeding brilliantly at the wrong objective.
When to use: You're leading AI strategy, digital transformation, or agent deployment and you need a structured diagnosis of why your AI investments aren't delivering expected value. Or you're preparing a case for leadership about what's actually missing.
What you'll get: A three-layer maturity assessment, a risk map of your most vulnerable AI deployments, a "Klarna test" for your highest-stakes agent, and a prioritized investment roadmap.
What the AI will ask you: Your industry, organizational size, current AI deployments, how organizational goals are communicated to AI systems, what's working, what isn't, and what keeps you up at night.
Prompt 4: Agent Intent Specification Generator
Job: Takes a specific AI agent or autonomous workflow and generates a complete intent specification — the machine-readable document that encodes what the agent should optimize for, what decisions it can make autonomously, when to escalate, how to resolve tradeoffs, and how to measure alignment.
When to use: You're deploying (or have already deployed) an agent and need to build the intent layer the article describes. This is the construction prompt — it builds the thing that would have prevented Klarna's failure.
What you'll get: A structured intent specification document with goal decomposition, decision boundary matrix, escalation triggers, value hierarchy, and feedback loop design — ready to be translated into system prompts, guardrails, or agent configuration.
What the AI will ask you: What the agent does, what organizational goal it serves, what decisions it makes, what tradeoffs it encounters, what "success" and "failure" look like, and what your most experienced human employee knows intuitively that the agent doesn't.
Prompt 5: AI Workflow Capability Map
Job: Maps your team's or organization's workflows into three categories — agent-ready (fully autonomous), agent-augmented (human-in-the-loop), and human-only — with the intent requirements, context needs, and decision authority levels for each.
When to use: You need to move from ad hoc AI adoption to systematic workflow architecture. You want to know where to invest in automation, where to invest in augmentation, and where to protect human judgment — and what intent infrastructure each category requires.
What you'll get: A complete workflow capability map with categorization, intent requirements, context dependencies, risk assessments, and an implementation sequence.
What the AI will ask you: Your team/department function, the key workflows your team performs, current AI usage, organizational risk tolerance, and what judgment calls require human involvement.