From mozilla-bigquery-etl-skills
ALWAYS use this skill when users ask to "create a skill", "make a skill for...", "add a new skill", or similar requests. This skill guides the creation of effective skills in the bigquery-etl repository that extend Claude's capabilities with specialized knowledge for BigQuery ETL workflows, Mozilla data platform conventions, or telemetry analysis. CRITICAL - First checks for conflicts with existing skills and recommends using/updating existing skills when appropriate. Do NOT attempt to create skills without invoking this skill first.
npx claudepluginhub mozilla/bigquery-etl-skills --plugin bigquery-etl-skillsThis skill uses the workspace's default tool permissions.
This skill provides guidance for creating effective skills tailored to the Mozilla bigquery-etl repository.
Creates isolated Git worktrees for feature branches with prioritized directory selection, gitignore safety checks, auto project setup for Node/Python/Rust/Go, and baseline verification.
Executes implementation plans in current session by dispatching fresh subagents per independent task, with two-stage reviews: spec compliance then code quality.
Dispatches parallel agents to independently tackle 2+ tasks like separate test failures or subsystems without shared state or dependencies.
This skill provides guidance for creating effective skills tailored to the Mozilla bigquery-etl repository.
🚨 CRITICAL FIRST STEP: Before creating any new skill, this skill checks for conflicts with existing skills and recommends using or updating existing skills when appropriate. See Step 0 below.
Skills are modular packages that extend Claude's capabilities with specialized knowledge and workflows. In bigquery-etl, they transform Claude into a specialized data engineer with Mozilla-specific expertise.
Every skill consists of:
skill-name/
├── SKILL.md (required) - Keep under 500 lines
│ ├── YAML frontmatter (name + description)
│ └── Markdown instructions
└── Bundled Resources (optional)
├── scripts/ - Executable code for deterministic operations
├── references/ - Documentation loaded as needed
└── assets/ - Templates and examples for output
When to include bundled resources:
CRITICAL: Before creating a new skill, check if an existing skill already covers this functionality.
Review existing skills to understand their scope:
Analyze the requested functionality:
Check for conflicts or overlap:
Inform the user of findings:
Before creating a new skill, I checked the existing skills:
[If conflict found]
The functionality you described overlaps significantly with the [existing-skill] skill, which handles [description].
You have two options:
1. **Use the existing skill**: [Brief guidance on how to use it]
2. **Update the existing skill**: If the current skill doesn't fully meet your needs, we can enhance it by [specific improvements]
Which would you prefer?
[If no conflict]
This appears to be a new capability that doesn't conflict with existing skills. Let's proceed with creating a new skill.
Only proceed to Step 1 if:
Skip if usage patterns are already clear.
Ask user for concrete examples of how the skill will be used:
Analyze each example to identify helpful resources:
Skip if skill already exists.
New skills belong in the bigquery-etl-skills repository, not in bigquery-etl. From the bigquery-etl-skills repo root:
mkdir -p skills/<skill-name>/{scripts,references,assets}
After creating files, register the skill in .claude-plugin/marketplace.json by adding "./skills/<skill-name>" to the skills array — without this the plugin will not load the new skill.
Create SKILL.md with frontmatter:
---
name: skill-name
description: Detailed description of when and how to use this skill for BigQuery ETL tasks.
---
Create bundled resources first (scripts, references, assets identified in Step 2).
Then write SKILL.md:
Answer these questions:
For detailed guidance:
references/explicit_instructions_pattern.md for how to ensure Claude reads assets and referencesStructure:
Content:
Resources:
After testing, users may request improvements:
Common improvements:
IMPORTANT: Always check these before creating a new skill (see Step 0).
Foundation:
Construction workflows:
Schema workflows:
Discovery:
Deployment:
Meta:
file.md" instructions in workflowsFor detailed guidance on specific topics, read these reference files:
references/mozilla_ecosystem.md - Mozilla data platform tools, repositories, APIs, and configuration files for construction-focused skills beyond bigquery-etlreferences/skill_composition_patterns.md - How to create composable skills that work together, with examples and anti-patternsreferences/explicit_instructions_pattern.md - How to ensure Claude reads assets and references using standardized sections and clear directivesreferences/potential_script_opportunities.md - When to use scripts in skills and high-priority script opportunities for existing skills