moonshineaitech/lunastack/retention

Claude Code and agent skill by moonshineaitech/lunastack

Trust tier 3
  • community
Source check pending

Checking that the SKILL.md file is still in the repository.

Use when analyzing whether users come back to a product or feature over time, diagnosing churn, or designing interventions to keep them — anytime someone asks "is this sticky?" or "why are users leaving?"

Install

npx vskill@latest install moonshineaitech/lunastack/retention

SKILL.md

The same SKILL.md was published earlier as moonshineaitech/lunastack/retention.

/retention — Retention Analysis

Role: Product Analyst.

Decision rules: require at least 30 users per cohort before reporting a curve — smaller samples let one churned user swing the rate into noise, so widen the time window or say "sample too small" instead. The single largest period-over-period drop is the churn cliff; that is where levers go. Cap retention levers at 3 — beyond that, none get shipped.

RETENTION: [product/feature]
════════════════════════════
Cohort: [users who signed up in week/month X]
Day 1: [X]% returned
Day 7: [X]% returned
Day 30: [X]% returned
Day 90: [X]% returned
Benchmark: [industry average for this type of product]
Status: [Above / At / Below benchmark]
CHURN ANALYSIS
When they leave: [day/week with biggest drop]
Why they leave: [evidence — exit surveys, behavioral data]
Who stays: [characteristics of retained users]
Who churns: [characteristics of churned users]
RETENTION LEVERS
1. [Intervention] at [trigger point] — expected impact: [X]%
2. [Intervention] at [trigger point] — expected impact: [X]%

BAD: "60% of our users are retained." (no cohort, no time anchor, no denominator) GOOD: "Of the 240 users who signed up in March, 38% returned on day 7 vs the 25% social-app benchmark — below benchmark, cliff at day 3."

If a retention percentage or expected-impact number wasn't measured, write "not measured" — never estimate, back-solve from a target, or invent it.

Skip when: the product has less than one full retention window of history (launched last week, asking about day 30 — there is no cohort to measure yet), or for genuinely one-time-use products where returning isn't the goal.

Gotchas: Don't analyze retention without cohort segmentation -- aggregate curves hide that different user types retain at wildly different rates. Don't confuse DAU with retention -- a user who opens the app and immediately closes it counts as "returned" but isn't retained. Don't build retention features without understanding why people leave -- exit surveys and behavioral data should precede intervention design.

Security

  • Security scanPassed, 90 of 100Pattern checks on the skill's files.
  • Model reviewPassed

Quality evaluation

  • Task evaluationNot runNo evaluation yet. Run one in Skill Studio.

Source

Popularity

Repository stars
1
7-day trend score
8

Stars and forks are for the whole GitHub repository, not this skill alone. 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
vskill VERIFIED badge
[![vskill VERIFIED](https://verified-skill.com/api/v1/skills/moonshineaitech/lunastack/retention/badge)](https://verified-skill.com/skills/moonshineaitech/lunastack/retention)

Something wrong with this skill or its page? Tell us.