menu

What If Your Repository Could Explain Itself?

Leader ●3 ●6 ●33
calendar_today • schedule1 min read

🚀 Built: GitLab Knowledge Navigator

Over the last week, I've been building GitLab Knowledge Navigator for the GitLab AI Hackathon 2026.

The idea started from a common developer problem:

You join a project, open an issue, and suddenly need to understand years of architecture decisions, merge requests, dependencies, ownership history, and code relationships.

The information exists, but it's scattered everywhere.

GitLab Knowledge Navigator transforms GitLab Orbit's knowledge graph into actionable repository intelligence.

Demo Video:- https://youtu.be/xLBaZ2sFMEI

🔍 What it can do:

• New Contributor Onboarding
• Architecture Memory
• Decision Intelligence
• Change Impact Analysis
• Architecture Drift Detection
• Repository Risk Alerts
• Executive Engineering Reports
• Knowledge Graph Visualization
• Repository Intelligence Scoring

One of the most interesting capabilities is the ability to answer questions such as:

"Why does this dependency exist?"

Instead of giving a generic AI answer, Navigator traces real repository history through GitLab Orbit and explains when a technology was introduced, why it was added, and which merge request introduced it.

🏗 Technical Highlights

• GitLab Orbit API
• Orbit Local Graph (DuckDB)
• Gemini 2.5 Flash
• Python
• Knowledge Graph Traversal
• Multi-Repository Analysis
• Interactive HTML Dashboards

The project was validated not only on its own codebase but also on a fork of GitLab Runner containing:

✅ 1,366 files
✅ 9,314 definitions
✅ 58,340 indexed relationships

One of the biggest lessons from this project was that repository intelligence requires more than an LLM. The real value comes from combining AI with structured repository knowledge and software development lifecycle context.

From Code → Context → Decisions.

GitLab #GitLabOrbit #GitLabAIHackathon #DeveloperTools #SoftwareEngineering #AI #KnowledgeGraph #Python #OpenSource #DevTools #GenerativeAI

8 Comments

•

This is an awesome tool you have here, am sure it would solve need to know for developers, those joining or those taking over a project. nice work on this. Share the project let us take it for a spin.

2 votes
•

@[Gift Balogun] Thank you! I really appreciate the encouragement.

The project is currently part of an active hackathon submission, so I'm keeping the repository private until the judging process is complete. Once the results are announced, I plan to share the project publicly along with the source code, documentation, and demo materials.

The goal is exactly what you described: helping developers quickly understand unfamiliar codebases, architectural decisions, dependencies, and ownership information without spending days digging through repositories.

I'll definitely share an update once the hackathon concludes. Thanks again for the support and interest!

0
•

Why does this dependency exist?" traced through real repository history — that's exactly the right framing. Most tools answer "what is this?" The harder and more valuable question is always "how did it get here, and what decisions created it?" We apply the same logic to on-chain deployer wallets: not just what a wallet looks like now, but the full history of actions that led to its current state. From Code → Context → Decisions maps cleanly to Address → History → Intent. Solid architecture.

1 vote
•

@[VeritasLab] Thank you! That's exactly the problem I wanted to solve.

Most developer tools are very good at answering "what is this?" but much weaker at answering "why does this exist?" and "what decisions led here?". GitLab Orbit's graph made it possible to connect code, merge requests, issues, ownership, and history into a single context layer.

I really like the comparison with on-chain analysis. "Address → History → Intent" is very similar to the mental model behind Navigator's "Code → Context → Decisions". In both cases, understanding the current state requires understanding the chain of events that created it.

Appreciate the thoughtful feedback!

1
•

@[MorningStar47] Exactly — "the current state is a compressed summary of everything that led to it" applies whether you're reading a codebase or a wallet. The chain of decisions is the real data. Good luck with the hackathon.

1
•

@[VeritasLab] That's a great way of putting it.

"The current state is a compressed summary of everything that led to it" captures the idea better than I could. One of the motivations behind Navigator was that codebases often preserve the result of decisions but lose the context around them over time.

I wanted to use GitLab Orbit to reconnect code with its history—issues, merge requests, ownership, and architectural decisions—so developers can understand not just what exists today, but the path that led there.

I really like the "Address → History → Intent" and "Code → Context → Decisions" parallel. Thanks for the thoughtful perspective and the good wishes!

1
•

This is a great example of where AI becomes genuinely useful for engineering teams. Most repository knowledge isn't missing, it's fragmented across commits, MRs, issues and documentation. Turning that history into explainable context instead of generic AI summaries solves a real onboarding and maintenance problem. The "Why does this dependency exist?" use case is particularly compelling.

1 vote
•

@[reetainraina] Thank you! That's exactly the insight that motivated the project.

In most repositories, the knowledge isn't actually lost—it's distributed across merge requests, issues, commits, discussions, and code relationships. The challenge is connecting those pieces quickly enough for developers to use them when they need them.

One of the goals of GitLab Knowledge Navigator was to move beyond generic AI explanations and ground answers in repository history and context. The "Why does this dependency exist?" workflow became one of my favorite examples because it shifts the conversation from understanding code to understanding the decisions behind the code.

I appreciate the thoughtful feedback. It's encouraging to hear that the value of turning repository history into explainable context resonates with other engineers.

1
🔥 Join developers growing publicly
Share your knowledge, build in public, and grow your developer presence with a global community.
Join CoderLegion

More Posts

Your AI Doesn't Just Write Tests. It Runs Them Too.

Kevin Martinez - May 12

Dashboard Operasional Armada Rental Mobil dengan Python + FastAPI

Masbadar - Mar 12

I’m a Senior Dev and I’ve Forgotten How to Think Without a Prompt

Karol Modelskiverified - Mar 19

I Stopped Fixing Bugs Myself and Built an AI to Do It Instead

rkchellah - May 3

The Sovereign Vault — A Comprehensive Guide to Protocol-Driven AI

Ken W. Algerverified - Jun 4
chevron_left
Surendra Kumar
2.9k Points • 42 Badges
India • portfolio.gfood.in
12Posts
20Comments
4Connections
Flutter and Firebase developer from Banda, India. I spend my time building
real, production-grade m... Show more

More From MorningStar47

Overengineering Doesn't Always Win Hackathons — I Learned This the Hard Way

Why "Just Build JARVIS" Is Harder Than It Sounds (And What Actually Gets You Close)

Built an Autonomous DFIR Agent SIFT-AEGIS — Here's What I Learned

Related Jobs

View all jobs →

Commenters (This Week)

3 comments
1 comment
1 comment

Contribute meaningful comments to climb the leaderboard and earn badges!