Skip to content
SkillModel-led5,063 tokens in use

attribution

Works out which marketing drives revenue: reads over-claiming platforms, compares models, plans holdout tests and builds first-party attribution.

61,709 stars on the repository MIT Licensecoreyhaines31/marketingskills · skills/attribution
npx skills add coreyhaines31/marketingskills --skill attribution
View source

Install command as written in the repository's README

Skill label

measured from its files

What installing attribution costs

98tokens idle
5,063tokens in use
15,483tokens, up to

Idle is what sits in context while the skill is installed. In use is its instructions, loaded when it runs. Up to adds the files its instructions link. Counts are approximate, made offline with cl100k_base.

Model dependence Model-led

Script-ledMixedModel-led

Mostly instructions, so how well it works depends on the model that follows them.

This describes the skill's structure, how much of it is scripts. It does not say how well it works.

Touches

Nothing detected: no shell, network, file writes, git or credentials in its files.

Declared tags come from the skill's frontmatter; detected ones from its scripts, its code blocks and the commands it names in inline code. This is detected, not audited.

How it scores

Quality

How well the skill is written, counted from its own files: its frontmatter, its steps, examples, stated output and constraints, its size and whether every file it ships is one it names.How we score it →
Excellent
  • Frontmatter12/15
  • Structure19/25
  • Examples12/15
  • Output15/15
  • Size13/15
  • Resources15/15

Security

Starts at 100 and loses points for each thing in the skill's files that a reader should look at first.How we score it →
Excellent

No findings.

Repository health

Pre-computed score based on development activity, maintenance, community, maturity, and trend momentum.How we score it →
Excellent
  • Development activity100
  • Maintenance100
  • Community76
  • Maturity32
  • Momentum40

attribution has two halves. Interpretation helps judge which numbers to trust: platform claims, model comparisons, MMM, incrementality and holdout tests, and self-reported attribution for podcasts and dark social.

The build half helps you own your attribution: anonymous ids merged to a known person with identify, first-touch capture, and stitching conversions that finish on a third-party domain through metadata passthrough and webhooks.

What You Get

  • A written answer that names the source of truth and what each source says
  • A model comparison for long sales cycles
  • A confidence and gaps section that shows where the data is weak
  • A first-party design using identify and merge with guardrails
  • A plan for writing source, confidence and basis onto CRM records

Common Use Cases

  • Judging whether a channel really works when its dashboard over-claims
  • Planning an incrementality or holdout test
  • Crediting podcasts and dark social with self-reported attribution
  • Stitching a booking made on a third-party scheduling page back to the visitor

What Makes It Unique

It pairs interpretation with a build track and sets guardrails for the build: only the anonymous id is passed to third-party links, and referrers such as checkout and login pages are excluded from first-touch data. References cover attribution models, business types, first-party tracking and measurement paradigms.

Under The Hood

The SKILL.md and four references are loaded as context. The skill ships no scripts and runs nothing itself. Its build guidance describes code you would add to your own site and webhook handler, using tools such as PostHog, GA4 or Segment; those need your own accounts.

Join founders building with open source

Opinionated takes, migration guides, cost-saving tips, and insights from the open source ecosystem.

Subscribe on SubstackJoin 750+ subscribers