kienbui1995/magic-powers/ai-product-retention
Claude Code and agent skill by kienbui1995/magic-powers
- community
Checking that the SKILL.md file is still in the repository.
Use when designing AI products for long-term retention — stickiness patterns, daily engagement hooks, workflow integration depth, habit loops specific to AI, and measuring whether users actually keep using your AI feature.
Install
npx vskill@latest install kienbui1995/magic-powers/ai-product-retentionSKILL.md
The same SKILL.md was published earlier as kienbui1995/magic-powers/ai-product-retention.
AI Product Retention
When to Use
- Month-1 retention below 50% and don't know why
- Users try product once and never come back
- Designing onboarding to maximize activation
- Choosing between product directions (pick the stickier one)
- Building AI features that become indispensable
Core Jobs
1. The AI Retention Crisis
Only 5% of AI tools achieve meaningful retention (>50% month-12). Compare:
| Product | Month-12 Retention | Why | |
|---|---|---|---|
| GitHub Copilot | 80% | Daily coding habit, workflow integration | |
| ChatGPT Plus | 71% | Daily use, replaces multiple tools | |
| Google Gemini | 25% | Redundant with free alternatives | |
| Most new AI tools | <30% | One-time use, low frequency problem |
The retention formula: Frequency × Depth × Irreplaceability
2. Stickiness Design Framework
Four dimensions that predict retention:
Scoring: Rate 1-3 on each dimension. Total <6 = retention risk. Target 8-12.
3. Habit Loop Design for AI Products
B.J. Fogg's model applied to AI tools:
4. Activation Optimization (The Critical First Session)
Users who don't activate in session 1 almost never return:
5. Retention Metrics for AI Products
Beyond standard cohort retention:
Key Concepts
- Activation event — the specific action that predicts long-term retention; find it by data analysis
- Stickiness score — 1-12 rating across frequency/depth/personalization/switching cost
- Habit loop — trigger → action → reward → investment cycle that makes tool indispensable
- AI acceptance rate — % of AI outputs users actually use; primary quality metric for AI products
- Critical path integration — product is embedded in user's core workflow; highest retention
- Day 1 churn — users who don't return after first session; usually 60-80% for new products
Checklist
- [ ] Problem frequency is ≥weekly (daily = best)?
- [ ] Activation event identified and tracked?
- [ ] Activation rate >40% in first session?
- [ ] Habit loop designed: what brings user back tomorrow?
- [ ] AI output improves over time with user data (investment loop)?
- [ ] Switching cost designed in (user loses something valuable by leaving)?
- [ ] AI-specific metrics tracked (acceptance rate, dependency score)?
Key Outputs
- Stickiness score: 1-12 with dimension breakdown and improvement plan
- Activation funnel: steps to aha moment, activation rate, drop-off points
- Habit loop design: trigger / action / reward / investment for your product
- Retention dashboard: day 1/7/30 + AI-specific metrics (acceptance rate, dependency)
Output Format
- 🔴 Critical — monthly-frequency problem (retention ceiling is low), no activation event tracked, no mechanism to bring users back tomorrow
- 🟡 Warning — stickiness score <6, activation taking >30 minutes, no habit loop designed, generic AI outputs (no personalization)
- 🟢 Suggestion — add daily trigger notification, build user context that improves AI over time, measure acceptance rate to track quality
Anti-Patterns
- Building for low-frequency problems (quarterly reports can't retain weekly habits)
- Complex onboarding before showing value (users leave before aha moment)
- Stateless AI (no memory = no personalization = no switching cost)
- Measuring "logged in" as active (measure actual AI usage, not visits)
- Ignoring day 1 experience (80% of lifetime value decisions made in first session)
Integration
- Use with
ai-product-designfor streaming/UX patterns that support habit loops - Use with
llm-observabilityto track AI acceptance rate and quality signals - Use with
ai-product-validation(validate problem frequency before building) - Agent:
@solo-ai-builderreviews retention design before launch
Security
- Security scanPassed, 95 of 100Pattern checks on the skill's files.
- Model reviewPassed
Quality evaluation
- Task evaluationNot runNo evaluation yet. Run one in Skill Studio.
Source
- SKILL.md in the repositoryCheckingskills/ai-product-retention/SKILL.md
Popularity
- 7-day trend score
- 0
Not refreshed yet.
Works with
Any agent vskill installs to: 56 targets in all. These 39 have a page on this site.
- Claude Code
- Cursor
- GitHub Copilot
- Windsurf
- Codex
- Gemini CLI
- Cline
- Amp
- Roo Code
- Goose
- Continue
- Kilo Code
- Junie
- Trae
- OpenHands
- Qwen Code
- AI Agent
- Tabnine Chat
- Sourcegraph Cody
- Amazon Q
- Firebase Genkit
- Sweep
- Devin
- Phind
- Mentat
- Claude Desktop
- Aider
- GPT Pilot
- Plandex
- bolt.new
- v0
- Replit Agent
- Anse
- TypingMind
- BetterChatGPT
- LibreChat
- Warp AI
- Pieces
- Fig AI
Embed badge
[](https://verified-skill.com/skills/kienbui1995/magic-powers/ai-product-retention)