Skill

marketo-parser

Parse and validate Marketo Email 2.0 templates. Extract modules, variables, and structural information for large template files that exceed token limits. Includes comprehensive Marketo Email 2.0 syntax reference for answering questions about standards, validation rules, and best practices. Use for reading templates piece-by-piece, validating structure, documenting template usage, and learning Marketo syntax.

Invocation

How this skill is triggered — by the user, by Claude, or both

Slash command

/allGood-marketing:marketo-template-parser

User invocable

Model invocable

Inline context

Default effort

Context Preview

The summary Claude sees in its skill listing — used to decide when to auto-load this skill

This skill provides Python-based tools for parsing and validating Marketo Email 2.0 templates, plus comprehensive reference documentation. It solves the problem of working with large templates (4,000-5,000 lines) that exceed token limits by allowing you to:

Supporting Files

README.mdreferences/email-rendering-rules.mdreferences/getting-started.mdreferences/marketo-template-reference.mdreferences/validation_rules.mdreferences/variable-naming-conventions.mdrequirements.txtscripts/detect_version.pyscripts/generate_registry.pyscripts/get_module.pyscripts/lint_email.pyscripts/list_modules.pyscripts/list_variables.pyscripts/parser_utils.pyscripts/validate.pysetup.sh

SKILL.md

365 lines · ~3.4k tokens

Stats

LanguagePython

Parent stars0

MaintenanceExcellent

Last CommitMar 18, 2026

Actions

View Source View Plugin View on GitHub View README

Stats

Actions

Marketo Template Parser Skill

Overview

Read templates incrementally - Get module index, then extract specific modules one at a time
Validate template structure - Check against Marketo Email 2.0 standards
Extract variables - List all customization variables without reading entire file
Generate machine-readable registries - Produce JSON module registry for downstream tools
Answer Marketo questions - Reference complete syntax documentation for elements, attributes, and validation rules

For setup and usage guidance, see references/getting-started.md.

Prerequisites

Python 3 and beautifulsoup4. Install with: pip3 install beautifulsoup4

Available Scripts

1. list_modules.py - Get Module Index

Get a quick index of all modules (IDs, names, line numbers) as JSON.

python3 scripts/list_modules.py <template.html>
python3 scripts/list_modules.py <template.html> --no-line-numbers

Output: JSON array of modules

[
  { "id": "logo", "name": "Logo", "line": 496 },
  { "id": "hero1", "name": "Hero 1", "line": 941 }
]

Options:

--no-line-numbers - Exclude line numbers from output

2. get_module.py - Extract Specific Module

Extract the complete HTML content for one specific module, or a JSON summary.

python3 scripts/get_module.py <template.html> <module_id>
python3 scripts/get_module.py <template.html> <module_id> --no-styles
python3 scripts/get_module.py <template.html> <module_id> --summary

Options:

--no-styles - Strip <style> tags to reduce token usage
--summary - Return JSON summary instead of full HTML

Summary output includes editable element details and variable metadata:

{
  "id": "hero1",
  "name": "Hero 1",
  "line": 941,
  "add_by_default": false,
  "editable_elements": 3,
  "editable_types": ["cta", "mktoText"],
  "editable_details": [
    { "id": "hero1Btn", "type": "cta", "name": "Button" },
    { "id": "hero1Headline", "type": "mktoText", "name": "Headline" }
  ],
  "variable_references": ["bg_color_grey", "btn_bgcolor", ...],
  "variable_details": [
    { "id": "bg_color_grey", "type": "mktoColor", "default": "#f5f5f5" },
    { "id": "btn_bgcolor", "type": "mktoColor", "default": "#6837ef" }
  ],
  "html_size": 1200,
  "html_lines": 35
}

3. list_variables.py - List All Variables

Get all variable declarations from <head> as JSON.

python3 scripts/list_variables.py <template.html>
python3 scripts/list_variables.py <template.html> --type mktoColor
python3 scripts/list_variables.py <template.html> --scope global

Options:

--type <type> - Filter by variable type (mktoString, mktoColor, mktoBoolean, mktoNumber, mktoList, mktoHTML, mktoImg)
--scope <scope> - Filter by scope (global or module)

Output format:

{
  "global": [
    { "id": "global_preheader", "type": "mktoString", "name": "Preheader", "default": "", "line": 26 }
  ],
  "module": [
    { "id": "bg_color_white", "type": "mktoColor", "name": "BG Color", "default": "#FFFFFF", "line": 47 }
  ]
}

4. validate.py - Validate Template Structure

Validate template against Marketo Email 2.0 standards and identify structural issues.

python3 scripts/validate.py <template.html>

Validation checks:

Critical errors:

Exactly one mktoContainer element
All IDs are unique (no duplicates)
ID format valid (letters, numbers, dash, underscore only)
Required attributes present (mktoName on modules and editable elements)
{{my.token}} in HTML attributes outside editable regions (won't resolve via fullContent API)

Warnings:

Lowercase attribute names (mktoname vs mktoName)
Unused variables (declared but not referenced)
Missing default values
Variable IDs not in camelCase (e.g. bg_color → bgColor)
URL variable defaults missing protocol (https://)

Output format:

{
  "valid": true,
  "score": 65,
  "errors": [],
  "warnings": [
    { "type": "case_sensitivity", "message": "Found 'mktoname' on line 27 (should be 'mktoName')", "line": 27 }
  ],
  "stats": {
    "modules": 57,
    "variables": 79,
    "unique_ids": 341,
    "editable_elements": 170,
    "containers": 1
  }
}

Score: starts at 100, -10 per error, -1 per warning. Valid if no errors.

5. generate_registry.py - Generate JSON Module Registry

Pure data extraction script. Produces a machine-readable JSON registry mapping each module to its editable elements and variable references. Output schema matches email-variant-generator/module-registry.json.

python3 scripts/generate_registry.py <template.html>
# Writes: <template>-registry.json

Output schema:

{
  "<moduleId>": {
    "addByDefault": true,
    "elements": { "<htmlId>": "<elementType>" },
    "content_variables": ["<varName>", ...],
    "style_variables": ["<varName>", ...]
  }
}

addByDefault - whether the module appears on the canvas when creating a new email (true) or must be dragged in from the sidebar (false). Mirrors the mktoAddByDefault HTML attribute; Marketo defaults to true when omitted.
elements - the editable content areas (mktoText, mktoImg, cta, etc.). This is where most per-email customization happens.
content_variables - variables that typically need to be set per-email (background image URLs, button link targets, preheader text). Usually 0-1 per module.
style_variables - colors, padding, font sizes, border radii, layout utilities. All have sensible defaults; rarely changed per-email.

Built-in validation runs automatically after generation:

Module count matches list_modules.py output
Every module name from list_modules.py appears in the registry
Element counts per module match get_module.py --summary output
Prints a summary: "Validated: 57 modules, 170 elements, registry OK" or error details

6. detect_version.py - Detect Template Version

Auto-detect whether a template is Email 1.0 or 2.0. For v1.0 templates, lists all mktEditable regions, extracts {{my.Token}} references, and provides upgrade guidance for migrating to Email 2.0.

python3 scripts/detect_version.py <template.html>

Output for v1.0 templates includes:

version: "1.0" or "2.0"
regions: array of mktEditable sections with IDs, token references, and line numbers
tokens: all {{my.Token}} names used across the template
upgrade: step-by-step migration guidance and per-region v2.0 conversion suggestions

For v2.0 templates, returns {"version": "2.0"} and directs to list_modules.py / validate.py.

Run this first on any unknown template to determine which scripts to use next.

7. lint_email.py - Email HTML Linter

Lint any email HTML for rendering issues across Outlook, Gmail, Apple Mail, dark mode, and accessibility. Works on any email HTML (v1.0, v2.0, or plain email — not Marketo-specific). 40 rules in 10 categories.

python3 scripts/lint_email.py <email.html>
python3 scripts/lint_email.py <email.html> --category accessibility outlook
python3 scripts/lint_email.py <email.html> --category gmail darkmode

Categories: structure, images, styles, links, tables, accessibility, size, outlook, gmail, darkmode

Output: JSON with issues array, summary counts, and score (100 - 5/error - 2/warning).

For details on what each rule checks and why, see references/email-rendering-rules.md.

Workflow: Generate Reference Documentation

Use this workflow to produce human-readable reference docs for any Marketo template. The LLM drives the process, making the workflow template-agnostic.

Run list_modules.py to get the module index (JSON array of {id, name, line})
Run generate_registry.py to get the machine-readable registry JSON
For each module (or in batches), use a subagent (smaller model) that: a. Receives the full module HTML via get_module.py <id> b. Also receives the --summary data (elements, variables, add_by_default) c. Returns a structured description: what the module does, what's editable, what's decorative, typical use cases
Assemble the reference docs:
- <template>-reference.md: overview table (module ID, name, editable types, LLM-generated description, add_by_default)
- <template>-details.md: full specs per module (elements with htmlIds/types, variables with types/defaults, LLM-generated description)
Cross-validate: module count in reference.md = count in registry.json, all module names match between the three artifacts

Other Workflows

Read Large Template Piece-by-Piece

# Step 1: Get module index (small JSON output)
python3 scripts/list_modules.py template.html > modules.json

# Step 2: Read modules.json to see what's available

# Step 3: Extract specific module for analysis
python3 scripts/get_module.py template.html hero1 > hero1.html

# Step 4: Analyze hero1.html (only ~25 lines instead of 4,697)

Validate Template Quality

python3 scripts/validate.py template.html

# Get validation summary
python3 scripts/validate.py template.html | jq '{valid, score, errors: .errors | length, warnings: .warnings | length}'

Document Template Usage

# Get all modules
python3 scripts/list_modules.py template.html > modules.json

# Get all variables
python3 scripts/list_variables.py template.html > variables.json

# Get validation stats
python3 scripts/validate.py template.html > validation.json

# Claude reads these 3 small JSON files and generates markdown documentation

Find Specific Module Type

# List all modules and filter by name
python3 scripts/list_modules.py template.html | jq '.[] | select(.name | contains("Hero"))'

Common Questions & Reference Usage

"What attributes can I use on mktoImg elements?"

Read references/marketo-template-reference.md (Section 2.4 mktoImg)

"What modules are available in the template?"

Run python3 scripts/list_modules.py <template.html>

"What variables does the hero1 module use?"

Run python3 scripts/get_module.py <template.html> hero1 --summary

"Why is my template validation failing?"

Read references/validation_rules.md (validation criteria)

"Can I nest modules inside modules?"

Read references/marketo-template-reference.md (Section 3.1 - Critical Constraints)

"Will this email render correctly in Outlook/Gmail?"

Run python3 scripts/lint_email.py <email.html>

"What are the dark mode / accessibility issues?"

Run python3 scripts/lint_email.py <email.html> --category darkmode accessibility
Read references/email-rendering-rules.md for client-specific details

Reference Files

General Marketo Knowledge

references/marketo-template-reference.md - Complete Marketo Email 2.0 syntax reference

Full syntax documentation for all Marketo elements (mktoModule, mktoText, mktoImg, etc.)
Validation rules and critical constraints
Best practices and common patterns

references/validation_rules.md - Validation criteria for templates

Critical errors (container rules, ID uniqueness, required attributes, unresolved tokens)
Non-critical warnings (case sensitivity, unused variables, naming conventions)

references/variable-naming-conventions.md - Variable ID naming guide

camelCase convention with examples
URL variable rules (protocol in defaults, ${variableId} vs {{my.token}})
Why tokens in href attributes won't resolve via fullContent API, with fix options

references/email-rendering-rules.md - Email HTML rendering rules by client

Outlook Word engine constraints and workarounds (ghost tables, VML, padding rules)
Gmail sanitizer rules (8KB style limit, 102KB clip, background-image stripping)
Apple Mail dark mode opt-in, auto-link prevention
Dark mode client matrix (no-change / partial / full invert)
WCAG accessibility requirements for email
Responsive hybrid/spongy layout patterns

references/getting-started.md - Setup and usage guide

Template-Specific References (Generated)

When you generate a registry or reference docs for a template, you get:

<template>-registry.json - Machine-readable module registry

Maps each module ID to its elements and variables
Compatible with email-variant-generator/module-registry.json schema

<template>-reference.md - Quick overview (generated by LLM workflow)

Table of all modules with descriptions

<template>-details.md - Full specifications (generated by LLM workflow)

Complete variable lists per module with types and defaults

Notes

Case sensitivity: Marketo allows lowercase attributes (mktoname), but camelCase (mktoName) is recommended for consistency
Warnings vs Errors: Errors prevent template from working; warnings are best-practice violations
Line numbers: All scripts include line numbers for easy navigation to issues
JSON output: All scripts output JSON for easy parsing and filtering with jq

marketo-parser

Invocation

Context Preview

Supporting Files

SKILL.md

marketo-parser

Invocation

Context Preview

Supporting Files

SKILL.md

Marketo Template Parser Skill

Overview

Prerequisites

Available Scripts

1. list_modules.py - Get Module Index

2. get_module.py - Extract Specific Module

3. list_variables.py - List All Variables

4. validate.py - Validate Template Structure

5. generate_registry.py - Generate JSON Module Registry

6. detect_version.py - Detect Template Version

7. lint_email.py - Email HTML Linter

Workflow: Generate Reference Documentation

Other Workflows

Read Large Template Piece-by-Piece

Validate Template Quality

Document Template Usage

Find Specific Module Type

Common Questions & Reference Usage

Reference Files

General Marketo Knowledge

Template-Specific References (Generated)

Notes

Similar Skills

Marketo Template Parser Skill

Overview

Prerequisites

Available Scripts

1. list_modules.py - Get Module Index

2. get_module.py - Extract Specific Module

3. list_variables.py - List All Variables

4. validate.py - Validate Template Structure

5. generate_registry.py - Generate JSON Module Registry

6. detect_version.py - Detect Template Version

7. lint_email.py - Email HTML Linter

Workflow: Generate Reference Documentation

Other Workflows

Read Large Template Piece-by-Piece

Validate Template Quality

Document Template Usage

Find Specific Module Type

Common Questions & Reference Usage

Reference Files

General Marketo Knowledge

Template-Specific References (Generated)

Notes

Similar Skills