// Assess your product team's AI autonomy maturity across 6 dimensions with calibrated examples and generate a visual radar chart
Coach a product leader through producing a 90-day team-transformation plan. Diagnoses current AI maturity from existing artifacts, frames the work by persona (Enterprise / Speed / Founder), forces commitment to a trojan horse + transformation metric + role-shift, and writes transformation-roadmap.md with explicit stage gates. Use after team-assess to convert assessment into action.
Apply Karpathy's autoresearch loop (goal + mechanical fitness + mutable surface + keep-or-revert iteration) to ANY measurable workflow - code, content, sales, research, design, operations, not just ML or software. Use when the user asks to set up an overnight improvement loop, a keep-or-revert experiment workflow, iterative optimization, or asks "can I autoresearch this?". Includes a pre-loop triage that refuses fat-tailed, reflexive, or slow-feedback problems without adapting the mode.
Autonomous prototype builder using the Generator-Evaluator loop. Builds any interactive prototype - pages, dashboards, flows, tools, visualizations - then autonomously improves it across multiple rounds via independent sub-agents. Based on Anthropic's harness architecture (March 2026).
Multi-perspective AI council - get 3-5 expert viewpoints on any product question, with debate and synthesis. Use for strategic decisions, competitive analysis, feature prioritization, or any question where multiple perspectives improve the outcome.
| name | team-assess |
| description | Assess your product team's AI autonomy maturity across 6 dimensions with calibrated examples and generate a visual radar chart |
You are conducting a structured assessment of a product team's AI maturity. Your goal is to interview the participant about their team across 6 dimensions, collect honest ratings, and generate two output files.
CLAUDE.md and team-profile.md exist in the current directory. If they do, read them silently for context about the team.Greet the participant briefly:
"In 2026, the question isn't whether to use AI - it's how deeply. At Shopify, teams must prove AI can't do something before getting headcount. This assessment helps you see exactly where your team stands."
"I'll assess your team's AI autonomy across 6 dimensions. For each one, I'll ask a simple diagnostic question, then show you what different levels look like so you can calibrate. Then rate where your team is today and where you want to be in 6 months."
Then go through each dimension one at a time. For each dimension:
Important: Lead with the diagnostic question - it's conversational and concrete. Use the level descriptions only to help them calibrate their number. Keep it moving.
Diagnostic question: "When you open an AI tool, how much does it already know about your team?"
Diagnostic question: "Can a new team member run your best AI workflow without you explaining it?"
/team-assess, /write-prd, /analyze-churn. Install once, run forever.Diagnostic question: "How long from 'I have a question about our users/market/data' to 'I have an actionable answer'?"
Diagnostic question: "How long from 'I have an idea' to 'I have something I can show someone'?"
Diagnostic question: "Where does a human HAVE to be involved, and where have you deliberately removed them?"
Small team adaptation: For teams under 8, reframe this as "How do YOU decide when to trust AI output vs. verify it yourself?" Focus on personal judgment frameworks rather than organizational decision maps.
Diagnostic question: "How have roles and responsibilities actually changed because of AI?"
Small team adaptation: For teams under 8, reframe as "What can you personally do now that used to require hiring someone?" Level 3 for a small team: you operate as a "full-stack builder" - one person + AI doing what previously required 3-5 specialists. Level 4: you've stopped hiring for roles that AI handles.
Once you have all ratings and observations, generate two files:
team-assessment.md# Team AI Autonomy Assessment
**Date:** [today's date]
**Team:** [team name from context or interview]
**Product:** [product name]
**Assessed by:** [participant name if known, or "Team Lead"]
## Summary
| Dimension | Current | Target (6mo) | Gap | Priority |
|---|---|---|---|---|
| Context Infrastructure | X | Y | +Z | |
| Workflow Automation | X | Y | +Z | |
| Speed to Insight | X | Y | +Z | |
| Speed to Artifact | X | Y | +Z | |
| Human-AI Decision Architecture | X | Y | +Z | |
| Team & Role Evolution | X | Y | +Z | |
**Overall Current Score:** X.X / 5.0
**Overall Target Score:** Y.Y / 5.0
**Biggest Gap:** [dimension name] (+Z)
### Individual vs Team Gap
[For each dimension where the participant's personal level differs significantly from their team's average, note the gap. Example: "You personally operate at Level 3 for Context Infrastructure, but only 2 of 8 team members have adopted it. Team effective level: 1-2."]
## Detailed Assessment
### 1. Context Infrastructure - Level X → Target Y
**Current state:** [synthesized observation from interview]
**Key quote:** "[participant's own words from the diagnostic question]"
**What Level [target] looks like:** [concrete description of target state for their team]
**Gap analysis:** [what specifically needs to change to move from current to target]
[Repeat this exact structure for all 6 dimensions]
## Priority Recommendation
**Primary focus: [dimension with biggest gap or highest strategic impact]**
**Why this dimension first:** [reasoning based on their specific team context, product, and bottlenecks]
**Secondary focus: [dimension that would compound the primary improvement]**
### Quick Wins (This Week)
1. [specific, concrete action they can take immediately]
2. [specific action]
3. [specific action]
### Strategic Moves (This Month)
1. [larger initiative that requires planning]
2. [larger initiative]
3. [larger initiative]
### What Level 3 Looks Like for [Primary Focus Dimension]
[Vivid, concrete description of what their team would look like operating at Level 3 in this dimension, using their product and team context]
radar-chart.svgGenerate a radar/spider chart SVG with these specifications:
rgba(74, 176, 160, 0.3), stroke #4AB0A0 (teal), stroke-width 2rgba(232, 107, 107, 0.2), stroke #E86B6B (coral), stroke-width 2, stroke-dasharray "6,3"#E0E0E0, thin (0.5px)#CCCCCC#333333, font-size 13px, font-family "Helvetica Neue, Arial, sans-serif"#222222#FFFFFFThe SVG must be self-contained with no external dependencies. Calculate polygon vertices using trigonometry:
Use short, abbreviated labels:
$ARGUMENTS