Product Tracking Skills

Your product isn't data-ready. These skills fix that.

You're paying for Amplitude, Mixpanel, or PostHog — but you can't answer three basic questions: Which features do your customers actually use? Where do users drop off? Which accounts are at risk?

The problem isn't your analytics tool. It's your instrumentation.

Product Tracking Skills is a set of open-source AI agent skills that scan your codebase, design an opinionated tracking plan, and generate real instrumentation code — for any analytics SDK, in any AI agent tool.

Works in: Claude Code, Codex, VS Code, and any tool with AI agent support.

See It In Action

Open your codebase in any AI agent tool and start talking:

You: audit tracking
AI:  [Finds every tracking call, identifies gaps and issues]
     Found 14 events across 8 files. Saved to .telemetry/current-state.yaml

You: design tracking plan
AI:  [Designs best-practice tracking plan, produces delta from current state]
     22 events. Delta: add 10, rename 3, change 4, remove 1. Review and adjust.

You: implement tracking
AI:  [Generates typed wrapper functions, delivery infrastructure, event constants]
     Tracking code ready in tracking/

Seven skills. One session. Data-ready.

The Problem

Most B2B products have one of these situations:

No tracking. You know you need it. It's been on the backlog for six months. It never happens.

Broken tracking. 14 events across 23 files. Some camelCase, some snake_case. No account context. Three events that do the same thing. Five that nobody uses.

Decayed tracking. Someone set it up 18 months ago. Twelve features have shipped since. None were instrumented. The tracking plan — if one exists — is a lie.

In all three cases, your CS team can't see which accounts are healthy. Your product team can't measure feature adoption. You can't give investors real usage numbers.

How It Works

Seven skills, each producing artifacts that feed the next. Everything version-controlled in your repo.

Business Case → Model → Audit → Design → Instrument → Implement → Maintain

What's Inside

These aren't thin prompts. Each skill includes a built-in reference library:

• 19 common instrumentation mistakes and how to detect them
• Anti-pattern detection — PII in event properties, noise events, redundant tracking, blocking calls
• Naming conventions — dot.notation for event names, snake_case for properties and traits
• B2B entity modeling — two-entity model, group hierarchies, instance vs user-level tracking
• 7 category templates — opinionated starting points for B2B SaaS, AI/ML, dev tools, and more
• SDK guides for 6 analytics platforms with correct identify, group, and track call patterns

Supported Analytics SDKs

Code generation is SDK-specific. Not generic wrappers — real call patterns, identity management, and group handling for each platform.

Skills Reference

Each skill is self-contained. Trigger them with natural language.

Installation

CLI Install (Recommended)

npx skills add accoilhq/product-tracking-skills

Install specific skills only:

npx skills add accoilhq/product-tracking-skills --skill audit design

Claude Code Plugin

/plugin marketplace add accoilhq/product-tracking-skills
/plugin install product-tracking-skills

Clone and Copy

git clone https://github.com/accoilhq/product-tracking-skills.git
cp -r product-tracking-skills/skills/* .claude/skills/

Example Prompt

Once installed, open your codebase and try:

Use the product-tracking skills to model this product, audit the current tracking,
and design a complete tracking plan. Our tracking destination is Segment — use the
Segment library. Track at the account and user level. Give me a complete, product
analytics tracking plan.

This runs the full skill chain — model, audit, design — and produces a tracking plan in .telemetry/ ready for implementation.

What You Get

Everything lives in .telemetry/ in your repo. Version-controlled. Survives across sessions and engineers.

.telemetry/
├── business-case.md        # Why add telemetry — stakeholder-ready
├── product.md              # What your product does, entities, value flow
├── current-state.yaml      # What's tracked today (from audit)
├── tracking-plan.yaml      # What should be tracked
├── delta.md                # Current → target diff (the backlog)
├── instrument.md           # SDK-specific instrumentation guide
├── changelog.md            # How the plan evolved over time
└── audits/
    └── 2026-02-13.md       # Timestamped audit snapshots

Who Is This For?

Founders who know they need product analytics but it's been on the backlog for six months. Go from zero to a complete tracking plan with working code in a single session.

Product engineers who inherited tracking that's scattered, inconsistent, and undocumented. Audit what exists, design what should exist, generate typed code to close the gap.

CS and product teams who need feature adoption data and account health signals. Hand engineering a clear process: "Run these skills. Get us the data we need."