gitlab2md 0.1.0

Convert GitLab profile data to Markdown for LLM analysis

gitlab2md

Convert GitLab profile data to clean Markdown files - the ideal format for LLM analysis.

Why Markdown?

Markdown is the lingua franca of AI tools. Once your GitLab data is in .md format, you can:

• Upload to NotebookLM and have conversations about your development history
• Use Claude Projects to analyze patterns across your contributions
• Feed to Obsidian with AI plugins for a personal developer knowledge base
• Run local LLMs (Ollama, LM Studio) for completely private analysis

Example Prompts

Once your GitLab data is in an LLM, try asking:

Question	Data Used
"What technologies do I work with most?"	projects.md, merge_requests.md
"Summarize my contribution patterns"	events.md, merge_requests.md
"What open source projects have I contributed to?"	contributed_projects.md
"Analyze my merge request review patterns"	merge_requests.md
"What groups am I most active in?"	groups.md, memberships.md
"Generate a developer portfolio summary"	All files
"What are my most starred projects?"	starred.md

Installation

Recommended (using pipx - installs in isolated environment):

pipx install gitlab2md

Or with pip (in a virtual environment):

pip install gitlab2md

Note: On modern Linux systems (Debian, Ubuntu 23.04+, Fedora), use pipx to avoid the "externally-managed-environment" error.

Prerequisites

• GitLab CLI (glab) installed and authenticated
• Python 3.13+

Installing glab

# macOS
brew install glab

# Ubuntu/Debian
sudo apt install glab

# Or via pipx
pipx install glab

Then authenticate:

glab auth login

Usage

# Use authenticated user
gitlab2md

# Specify a username
gitlab2md juanmanueldaza

# Include group contributions
gitlab2md juanmanueldaza --groups colmena-project,another-group

# Custom output directory
gitlab2md juanmanueldaza -o ./my_gitlab_data

Then drag the output folder into your favorite AI tool.

Options

Option	Description	Default
username	GitLab username (optional, uses authenticated user if omitted)	-
-o, --output	Output directory	gitlab_export
--groups	Comma-separated list of groups to include contributions from	-

LLM Tools That Work Great With This

Tool	Type	Best For
NotebookLM	Cloud	Conversational analysis, audio summaries
Claude Projects	Cloud	Deep analysis, long context
ChatGPT	Cloud	General analysis, quick insights
Obsidian + AI plugins	Local	Personal knowledge base, linked notes
Open Notebook	Local/Cloud	16+ AI models, open source
Ollama	Local	Private, offline analysis

Output Files

Creates 13 markdown files in the output directory:

Profile & Projects

• profile.md - User profile (name, bio, job title, organization, website)
• projects.md - Owned projects with stars, forks, visibility, topics
• member_projects.md - Projects where user is a member (excludes own)
• contributed_projects.md - Projects user has contributed to

Activity & Contributions

• merge_requests.md - MR history with state, labels, and status breakdown
• issues.md - Issues created with state and labels
• events.md - Recent activity summary and action types
• groups.md - Group memberships and commit contributions

Network & Social

• starred.md - Starred projects
• memberships.md - Groups and projects with access levels
• snippets.md - Code snippets with visibility

Security

• ssh_keys.md - SSH key titles and expiration dates
• gpg_keys.md - GPG key IDs and creation dates

Architecture

This project follows SOLID principles for maintainability:

• Single Responsibility: Each parser/formatter handles one data section
• Open/Closed: Add new parsers via @register_parser decorator without modifying core code
• Liskov Substitution: All parsers implement the SectionParser protocol
• Interface Segregation: Focused protocols for extractors, parsers, formatters, and writers
• Dependency Inversion: Converter depends on abstractions, not implementations

Contributing

Contributions are welcome! Please see CONTRIBUTING.md for guidelines.

License

GPL-2.0 - see LICENSE for details.