solo-metrics-track
Skill18 starsChanged 4 months agoMITView on GitHub
Use when "set up metrics", "track KPIs", "analytics events", "what to count", "funnel analysis", "when to kill or scale", "success metrics", or need analytics plan. Do NOT use for SEO metrics (/seo-audit).
In this repository62 other files it gives its agents.
- CLAUDE.mdCLAUDE.md
- solo-android-releaseSkill
- solo-apple-app-iconSkill
- solo-auditSkill
- solo-boardSkill
- solo-buildSkill
- solo-community-outreachSkill
- solo-content-genSkill
- solo-creator-growthSkill
- solo-customer-finderSkill
- solo-deep-researchSkill
- solo-deploySkill
- solo-diagnoseSkill
- solo-domainSkill
- solo-factorySkill
- solo-github-outreachSkill
- solo-gmail-cliSkill
- solo-grillSkill
- solo-hackathonSkill
- solo-humanizeSkill
- solo-i18nSkill
- solo-index-youtubeSkill
- solo-initSkill
- solo-ios-devSkill
- solo-ios-releaseSkill
- solo-kaggle-gpuSkill
- solo-kindleSkill
- solo-knowledgeSkill
- solo-landing-genSkill
- solo-launchSkill
- solo-legalSkill
- solo-memory-auditSkill
- solo-mirrorSkill
- solo-model-shrinkSkill
- solo-music-reels-cutSkill
- solo-open-datasetSkill
- solo-paperSkill
- solo-pdf-printSkill
- solo-pipelineSkill
- solo-planSkill
- solo-redditSkill
- solo-researchSkill
- solo-retroSkill
- solo-reviewSkill
- solo-scaffoldSkill
- solo-secretsSkill
- solo-seo-auditSkill
- solo-seo-cliSkill
- solo-setupSkill
- solo-sgrSkill
- solo-skill-auditSkill
- solo-skill-shelfSkill
- solo-stem-splitSkill
- solo-streamSkill
- solo-supervoxSkill
- solo-swiftui-design-systemSkill
- solo-transcribeSkill
- solo-validateSkill
- solo-video-editSkill
- solo-video-promoSkill
- The first 60 shown.
The file
We split this skill into the rules it gives an agent
9 rules. About 0 are common, 5 are distinctive, and 4 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈5
Distinctive · estimate
This author's own choices. Worth a look.
4
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 22
If MCP tools are not available, fall back to Grep + Read.
- Distinctive: line 33
Read PRD for features, ICP, monetization model.
- Very specific: line 34
Read CLAUDE.md for stack (iOS/Web/both).
- Distinctive: line 35
If empty: ask via AskUserQuestion.
- Very specific: line 46
If MCP available: kb_search("analytics events funnel") for past funnels
- Distinctive: line 55
Before defining KPIs, write out:
- Very specific: line 63
Use thresholds from references/kpi-benchmarks.md as starting point.
- Very specific: line 125
No cookie, no localStorage, nothing returned to the browser, so no consent banner.
- Distinctive: line 138
Cause: Default thresholds are industry averages.
≈ “Common” is an estimate: at least 1 in 5 of the 372,429 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
/metrics-track
Set up a metrics tracking plan for a project. Defines the event funnel for superduper-analytics (a self-hosted, cookieless event counter on Cloudflare, one schema for web and apps), KPI benchmarks, and kill/iterate/scale decision thresholds based on lean startup principles.
MCP Tools (use if available)
kb_search(query)— find analytics patterns, past funnels, measured benchmarks
If MCP tools are not available, fall back to Grep + Read.
Methodology Reference
This skill implements metrics tracking based on lean startup principles:
- Relative metrics vs niche benchmarks — compare against your own trajectory, not vanity averages
- Kill/iterate/scale decision rules — data-driven thresholds for product decisions (see step 7 below)
Steps
- Parse project from
$ARGUMENTS.- Read PRD for features, ICP, monetization model.
- Read CLAUDE.md for stack (iOS/Web/both).
- If empty: ask via AskUserQuestion.
- Detect platform:
- Web → the
sda.jstag (one line in<head>) +sda("event_name", props, metrics)calls - iOS / macOS → the SuperDuperAnalytics SPM package,
Analytics.track(...) - Android, KMP, Python, server → no client exists; POST events to the ingest endpoint (contract in the repo README and
src/schema.ts),platformset toandroid/server - Web + app → register ONE source id for both, so a landing visit and an app launch are compared without a join. There is no per-person link between them, by design (see Notes)
- Web → the
- Check what is already wired:
seo productslists every registered source and whether its counter reports. A product with no counter reads as zero traffic for reasons unrelated to how it is doing- Grep the project for
sda.js,sda(,Analytics.track, and for leftovers (posthog,gtag(,googletagmanager). Legacy sites may still have Google Analytics;seo gareads it, but don't add it to new ones - If MCP available:
kb_search("analytics events funnel")for past funnels
Define event funnel based on PRD features:
Standard funnel:
Awareness → Acquisition → Activation → Revenue → Retention → ReferralSee
references/kpi-benchmarks.mdfor the full event table, KPI thresholds, code snippets, and A/B test template. Adapt events to the specific product. Keep it to about ten event names per product: it is a counter, not session replay.- Forced reasoning — metrics selection:
Before defining KPIs, write out:
- North Star Metric: The ONE number that matters most (e.g., "weekly active users who completed core action")
- Leading indicators: What predicts the North Star? (e.g., "activation rate D1")
- Lagging indicators: What confirms success? (e.g., "MRR", "retention D30")
- Vanity metrics to AVOID: (e.g., total downloads without activation)
Set KPI benchmarks per stage:
Use thresholds from
references/kpi-benchmarks.mdas starting point. Adjust for product type (B2B has lower volume/higher conversion, B2C mobile has higher volume/lower retention).- Define decision rules — use kill/iterate/scale framework from
references/kpi-benchmarks.md. Adapt thresholds to project-specific values from step 6. - Generate implementation snippets — use platform-specific examples from
references/kpi-benchmarks.md. Adapt event names and properties to the funnel defined in step 4. - A/B Test Analysis Template — copy template from
references/kpi-benchmarks.mdinto the metrics plan. Include decision rules (SHIP / KEEP / EXTEND / INCONCLUSIVE). - Write metrics plan to
docs/metrics-plan.md:
# Metrics Plan: {Project Name}
**Generated:** {YYYY-MM-DD}
**Platform:** {iOS / Web / Both}
**North Star:** {north star metric}
## Event Funnel
| Stage | Event | Properties |
|-------|-------|------------|
{event table from step 4}
## KPIs & Thresholds
| KPI | Target | Kill | Scale |
|-----|--------|------|-------|
{benchmark table from step 6}
## Decision Rules
{framework from step 7}
## Implementation
### superduper-analytics setup
- Source id: {id} (one commit to `registry/sources.yaml`; unregistered ids are refused with 403)
- Client: {sda.js tag / SuperDuperAnalytics SPM / POST to the ingest endpoint}
- Identity: anonymous only (web: edge-derived, rotates daily; apps: install-scoped UUID)
- Store side: App Store `ct` campaign tokens / Play `referrer` UTM on every store link
### Code Snippets
{snippets from step 8}
## A/B Test Template
{template from step 9}
## Reading it
- Traffic, people vs bots: `seo traffic --days 7`
- Which pages make people act: `seo conversion --days 28`
- Do app users come back: `seo retention --days 28` (apps only; the web id dies at midnight UTC by design)
- Installs, purchases, subscriptions per campaign: App Store Connect / Play Console, by `ct` / UTM
---
*Generated by /metrics-track. Implement events, then review weekly.*- Output summary — North Star metric, key thresholds, first event to implement, A/B template included.
Notes
- Event names:
object_verb, past tense, snake_case (page_viewed,photo_exported). Props are strings, metrics are numbers - No cookie, no
localStorage, nothing returned to the browser, so no consent banner. That stays true only while nothing else on the page sets one - No
identify(). Linking one visitor to one install needs a cross-site identifier, which is exactly what this setup refuses. The funnel is aggregate: landing visits and store clicks from the counter, installs and purchases from the store's campaign report - Retention is an app metric here. Web returning visitors are invisible by design
- Errors and crashes are not analytics: App Store Connect crashes + MetricKit, Play Console Android vitals, Workers Logs. No third-party telemetry SDK
- Review weekly, make the kill/iterate/scale decision monthly
Common Issues
Wrong platform detected
Cause: Project has both web and iOS indicators. Fix: Skill checks package manifests. If both exist, it generates cross-platform identity setup. Verify the detected platform in the output.
KPI thresholds too aggressive
Cause: Default thresholds are industry averages.
Fix: Adjust thresholds in docs/metrics-plan.md based on your niche. B2B typically has lower volume but higher conversion.
Counter not in project
Cause: Metrics plan generated but nothing sends events.
Fix: This skill generates the PLAN only. Web: add the sda.js tag (docs/install.md in https://github.com/fortunto2/superduper-analytics). iOS: add the SPM package https://github.com/fortunto2/superduperai-analytics-swift. Then register the source id and check seo products.
Events sent, nothing arrives
Cause: The source id is not in registry/sources.yaml (403), or the request came from an origin the registry does not list (dev server, preview deploy: dropped silently, by design).
Fix: Register the id, deploy, and test from the production host.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

