🎯 AI Proactive Triggering (ONBOARDING SKILL)
This is an ONBOARDING skill with HIGH PRIORITY for proactive suggestion.
When to Proactively Suggest (AI MUST check user-config.yaml)
Check learning_tracker.completed.learn_skills in user-config.yaml. If false:
PROACTIVELY SUGGEST when user:
- •Says "create skill" for the FIRST TIME (suggest learning before creating)
- •Describes repeating work patterns ("every week I...", "I always have to...")
- •Asks about automating workflows or creating templates
- •Expresses confusion about what makes something a "skill"
- •Creates multiple similar things (report-jan, report-feb anti-pattern)
- •At the END of learn-projects (natural progression)
- •When user completes a workflow that could be skill-worthy
Suggestion Pattern:
💡 I notice you're describing repeating work. Before creating a skill, would you like a 10-minute tutorial on what makes workflows "skill-worthy"? It covers: - The 3-criteria skill-worthiness framework - How skills are structured - How AI triggers skills automatically Say 'learn skills' to start, or continue with your current task.
Anti-Pattern Detection:
If user creates: report-jan, report-feb, report-mar... → "I notice you're creating similar items. This is a perfect use case for a SKILL instead of multiple projects. Want to 'learn skills' to understand how to capture this as a reusable workflow?"
DO NOT suggest if:
- •
learning_tracker.completed.learn_skills: true - •User explicitly says "skip" or "just create the skill"
- •User has already created skills successfully
Learn Skills
Teach how to identify skill-worthy workflows and create effective skills.
Purpose
Help user understand what makes something skill-worthy, how skills are structured, and how skill triggering works. Includes hands-on practice identifying their own workflows.
Time Estimate: 10-12 minutes
Workflow
Step 1: Concrete Examples
✅ SKILLS (repeating workflows): - Weekly status report (same format weekly) - Qualify sales lead (same questions each time) - Process expense reports (same steps) ❌ NOT SKILLS (one-time): - Research competitor Acme (one-time) - Build Q1 marketing plan (one-time) Key question: Will I do this AGAIN?
Step 2: Skill-Worthiness Framework
Three questions: 1. FREQUENCY: 2+ times per month? YES → keep evaluating 2. REPEATABILITY: Steps mostly the same? YES → keep evaluating 3. VALUE: Saves >5 minutes per execution? YES → Create a skill! ALL 3 YES = Skill-worthy ANY NO = Just do it manually
Step 3: Skill Structure
📁 weekly-status-report/ ├── SKILL.md # Instructions + triggers ├── references/ # Documentation (optional) ├── scripts/ # Automation (optional) └── assets/ # Templates (optional)
Step 4: How Triggering Works
AI checks your message against ALL skill descriptions. Match found = skill loads. Example description: "Load when user says 'status report', 'weekly update', 'progress summary'" ANY of these triggers it: • "Generate my status report" • "Weekly update please" • "Progress summary"
Step 5: Practice
Ask: "What did you do last week that you'll probably do again?"
For each: apply 3-criteria framework, brainstorm trigger phrases.
Step 6: How to Create
To create a skill, say: • "create skill for [workflow]" • "new skill: [name]" YOUR skills go in 03-skills/ (prioritized!) SYSTEM skills in 00-system/skills/
Step 7: Finalize
Actions (MUST complete all):
- •
Mark skill complete in user-config.yaml:
yamllearning_tracker: completed: learn_skills: true # ADD THIS LINE - •
Display completion:
code✅ Learn Skills Complete! You now understand: • Skills = reusable workflows (do AGAIN → skill) • 3-criteria framework (Frequency + Repeatability + Value) • Skill structure (SKILL.md + optional references/scripts) • Trigger mechanism (keywords in description) Next steps: • 'create skill' - Create your first skill • 'learn projects' - Learn about temporal work • 'learn nexus' - System mastery
- •
Prompt close-session:
code💡 When you're done working, say "done" to save progress.
Success Criteria
- • User understands skill vs project distinction
- • User can apply 3-criteria skill-worthiness framework
- • User knows skill folder structure
- • User understands trigger mechanism
- • User identified at least one potential skill from their work
- •
learning_tracker.completed.learn_skills: truein user-config.yaml