Detailed guide to each level, how to recognize it, and how to level up.
User: "Write a function to validate emails"
AI: *writes function*
User: "Make it also check for disposable domains"
AI: *rewrites function*
User: "Add tests"
AI: *writes tests*
User: "The tests are failing"
AI: "Let me fix that..."
User: "Now add it to the user service"
AI: *adds to service*
User: "It broke the existing tests"
...continues for 15+ turns
User: "I need to add email validation to our user registration.
We use TypeScript, Jest for testing, and we have a UserService
that handles registration. We should check format and block
disposable email domains."
AI: "I'll create the validation with tests..."
User: "Looks good, but can you also add logging?"
AI: *adds logging*
User: "Great, now integrate with UserService"
AI: *integrates*
User: "Tests are failing on CI"
AI: "Let me check the CI config..."
User: "Add email validation to user registration. Use writing-plans
to create a plan first."
AI: *invokes writing-plans*
*creates comprehensive plan*
"Here's the plan. Ready to execute?"
User: "Approved"
AI: *invokes executing-plans*
*implements with checkpoints*
*invokes verification-before-completion*
"Done. All tests pass. Here's the summary."
User: "Add email validation, rate limiting, and audit logging
to user registration. Use dispatching-parallel-agents
since these are independent."
AI: *invokes dispatching-parallel-agents*
*dispatches 3 parallel agents*
*each agent uses appropriate skills*
*aggregates results*
*invokes verification-before-completion*
"All 3 features implemented, tested, verified."
User: "We keep doing the same email validation pattern.
Create a skill for it using writing-skills."
AI: *invokes writing-skills*
*creates reusable email-validation skill*
*includes tests, documentation, examples*
"Created skill. Now any team member can just invoke
'email-validation' and get consistent implementation."
The shift: Stop treating AI like a search engine.
Practice:
Skill to learn: None specific — just better prompting
Measure: Did your turns drop from 15+ to 10?
The shift: Stop steering and start delegating.
Practice:
writing-plans to create a plan before implementing"Skills to learn:
writing-plans — structured planningexecuting-plans — checkpoint-based executionverification-before-completion — verify before doneMeasure: Did your turns drop from 10 to 5?
The shift: Stop doing one thing at a time.
Practice:
dispatching-parallel-agentssubagent-driven-developmentSkills to learn:
dispatching-parallel-agents — parallel worksubagent-driven-development — context managementusing-git-worktrees — isolated workspacesMeasure: Did your turns drop from 5 to 2?
The shift: Stop optimizing yourself, start optimizing the team.
Practice:
writing-skillsSkills to learn:
writing-skills — create reusable skillsMeasure: How many team members are using your skills?
| Aspect | L1 | L2 | L3 | L4 | L5 |
|---|---|---|---|---|---|
| Turns/task | 15+ | 8 | 4 | 2 | <1 |
| Planning | None | Ad-hoc | Structured | Parallel | Systematic |
| Skills used | 0 | 1-2 | 3-5 | 5+ | Creates new |
| Context | Polluted | Growing | Clean | Managed | Shared |
| Rework | 60% | 35% | 15% | 5% | 0% |
| Impact | Self | Self | Self | Self | Team |
| Level | Current | Target | Change |
|---|---|---|---|
| L1 | 40% | 10% | -30% |
| L2 | 30% | 30% | 0% |
| L3 | 20% | 40% | +20% |
| L4 | 8% | 15% | +7% |
| L5 | 2% | 5% | +3% |
Moving 30% of users from L1 to L3:
This is the highest ROI investment in AI fluency.
Take the self-assessment to find your level, then focus on the next level up.