Skip to content

CONTRIBUTING.md

Latest commit

 

History

History
99 lines (68 loc) · 2.76 KB

CONTRIBUTING.md

File metadata and controls

99 lines (68 loc) · 2.76 KB
Raw

Contributing

Thanks for your interest in improving the GTM pipeline skills. This guide covers how to contribute.

How to Contribute

  1. Fork the repo
  2. Create a branch from main for your change
  3. Make your changes — follow the conventions below
  4. Test with ./install.sh and invoke the skill you modified
  5. Submit a PR to main

PRs are reviewed before merging. The stable branch is the maintainer's production copy — don't target it.

What You Can Contribute

  • Skill improvements: Better instructions, clearer steps, edge case handling
  • New provider support: Add a data provider to an existing skill's provider table
  • Bug fixes: Incorrect field names, broken references, missing steps
  • Documentation: Architecture docs, examples, README improvements
  • New skills: Propose via an issue first — discuss scope before building

Conventions

Skill Files

  • Each skill lives in skills/gtm-{name}/SKILL.md
  • Follow the YAML frontmatter format: 
    ---
name: gtm-pipeline:{skill-name}
description: One-line description. Trigger phrases at the end.
---
  • Reference shared files with ~/.claude/skills/gtm-pipeline/_shared/conventions.md
  • Use snake_case for all CSV field names (see _shared/conventions.md for canonical names)

No Personal Data

Never commit:

  • API keys, session cookies, or secrets
  • PhantomBuster agent IDs (these go in _shared/local.md, which is gitignored)
  • Absolute paths (use ~/.claude/skills/... or $HOME)
  • Client names, domains, or real contact data
  • Your name or identifying information

If a skill needs user-specific config, reference _shared/local.md with a config key.

Shared Files

  • _shared/conventions.md — field naming, terminology, directory structure
  • _shared/phantombuster.md — PB API patterns and script templates
  • _shared/local.example.md — template for personal config (update if you add new config keys)

Examples

If your change affects the expected project structure, update examples/sample-client-gtm/ to match.

Testing Your Changes

# Install skills from your fork
./install.sh

# Open a project in Claude Code
claude .

# Invoke the skill you changed
/gtm-pipeline:{skill-name}

Verify:

  • The skill runs without referencing missing files
  • Field names match _shared/conventions.md
  • No hardcoded paths, IDs, or personal data
  • Existing skills that reference your changed skill still work

Reporting Issues

Open an issue with:

  • Which skill is affected
  • What you expected vs. what happened
  • Your provider setup (which API keys you have configured)

Code of Conduct

Be constructive. This project exists to make GTM pipelines accessible — contributions that support that goal are welcome.