Yakoub-ai

Agent Mmm — AI skill for Claude Code

AI community

Marketing Mix Model expert agent plugin for Claude Code - pymc-marketing v0.18.2+.

How to install Agent Mmm

This entry records only its repository, not the path inside it, so there is no exact command to give. Open Yakoub-ai/agent-mmm and copy the folder into ~/.claude/skills/, or the file into ~/.claude/agents/.

What Agent Mmm does

Marketing Mix Model expert agent plugin for Claude Code - pymc-marketing v0.18.2+.

Alternatives in AI

Pal Configuration Docs — for model and thinking mode options 11.3k ★
Oh My Claudecode — 32 specialized agents and 7 execution modes for Claude Code, with smart model routing and automatic paralleliz 10.9k ★
Codeburn — Free, local tool to track AI coding token usage and cost across 37 tools and agents (Claude Code, Cursor, Code 9.7k ★

README

Yakoub AI Plugins — Claude Code Marketplace

A Claude Code plugin marketplace for data science and marketing analytics.

Plugins

[agent-mmm](./plugins/agent-mmm/) — Marketing Mix Model framework

Expert MMM knowledge plus an executable pipeline: a framework-agnostic project spec, data preparation and auditing, causal specification, prior generation, model fitting, diagnostics, and stakeholder reporting.

Works across **pymc-marketing 1.1**, **Google Meridian 1.8** and **Meta Robyn 3.12** — pymc-marketing runs in-process, the others are generated as runnable code with their data contracts, and the plugin reports what each framework cannot express rather than dropping it silently.

The knowledge is organised around the things that actually make MMMs wrong: variable roles (a price index is not a media channel and email has no ROAS), baseline health (a negative baseline invalidates every ROAS below it), identifiability (collinear channels cannot be separated, whatever the model reports), and calibration (an MMM without an experiment is an argument from correlation).

**Agents** — `agent-mmm`, `mmm-modeler`, `mmm-diagnostician`, `mmm-improver`, `mmm-reporter`

**Commands** — `/mmm-intake`, `/mmm-intake-quick`, `/mmm-prepare-data`, `/mmm-analyze-data`, `/mmm-recommend-controls`, `/mmm-recommend-priors`, `/mmm-build`, `/mmm-fit`, `/mmm-diagnose`, `/mmm-improve`, `/mmm-report`, `/mmm-status`

**Skills** — project planning, intake, data engineering, data quality, channel semantics, causal design, baseline and trend, model building, experimentation and calibration, validation, diagnostics, attribution, budget optimisation, multi-geo panels, target units, stakeholder reporting, iterative improvement, greenfield vs brownfield, external factors, and reference guides for pymc-marketing, Meridian, Robyn and framework selection.