glabflow 0.1.0a10

Async-native Python library for bulk operations on self-hosted GitLab

glabflow

The fastest GitLab client for AI agents and high-volume automation, written in Python.

• ⚡ 10-100x faster than python-gitlab — async-first, connection pooling, keyset pagination
• 🤖 AI agent tools — Native tool definitions for agent frameworks
• 📊 100% REST coverage — 1,124 operations (vs ~300 in python-gitlab)
• 🌍 102 GraphQL operations (27 queries + 75 mutations) — DataLoader batching, query caching, schema introspection codegen
• 📦 Single tool — replaces python-gitlab, gitlab-cli, and custom scripts

Install: uv add glabflow or pip install glabflow — Python 3.14+

---

Highlights

• AI agents — Native tool definitions for agent frameworks
• CLI included — gl covers auth, listing, git ops, audits, analytics, and more (see CLI below)
• Compliance — SOC2 Type II + ISO 27001 evidence collection, 9 controls mapped
• Config — Per-profile timeout, SSL verification, system-level (/etc/glabflow/config.json)
• Auth — OAuth2, JOB-TOKEN, HTTP Basic, anonymous access
• Error handling — on_http_error decorator for custom exception handling per-call

---

CLI (Command Line Interface)

glabflow ships a gl CLI for bulk GitLab operations — install with uv add glabflow[cli] or pip install glabflow[cli].

Command	Purpose
gl auth	Login, profiles, credential resolution
gl list	List repositories, users, or groups
gl git clone|pull|fetch	Bulk git operations across projects
gl users	Privilege checks, GraphQL-based activity/contribution log
gl mr	List merge requests, list commits in a merge request
gl pipelines	List, logs, stats, live-watch, CI failure-rate analysis
gl audit	Compliance dashboards, public/stale repo audits
gl analytics	DORA metrics, velocity, code churn, timelogs
gl spamcheck / gl anomaly	Spam user detection, access anomaly detection
gl upgrade	Pre-upgrade checklist, version diff
gl migrate	Migrate python-gitlab code to glabflow automatically
gl runner	Register and start a local GitLab Runner

Run gl --help or any subcommand with --help for full usage; see the CLI Reference for details.

---

Performance

Mode	Users/sec	vs python-gitlab
glabflow (GIL off)	1207	100-200x
glabflow (GIL on)	713	50-100x
python-gitlab	60-80	baseline

---

glabflow vs python-gitlab

Feature	glabflow	python-gitlab
AI agent tools	✅ First-party	❌
Speed	10-100x faster	Sequential
REST coverage	100% (1,124 ops)	~300 ops
GraphQL	102 ops + DataLoader + schema codegen	Experimental
Async	Native	Via python-gitlab[async]

---

Requirements

• Python 3.14+ (free-threaded recommended)
• aiohttp>=3.10, msgspec>=0.18, stamina>=24.2

Extras: uv add glabflow[cli,extended] or pip install glabflow[cli,extended]

---

Who's Using glabflow?

Production users include: AI automation platforms, Platform Engineering teams, DevOps teams, compliance tooling, and high-volume GitLab deployments.

Add your project via PR — we'd love to feature you.

---

Community

Built by developers who needed speed. Used by teams processing millions of GitLab API calls daily.

• AI automation platforms building agent-based workflows
• DevOps teams managing hundreds of GitLab instances
• Security teams running compliance audits at scale
• Platforms integrating GitLab into developer workflows

Join us on GitHub Discussions.