1st1

1st1 / lat.md

Public

Agent Lattice: a knowledge graph for your codebase, written in markdown.

www.lat.md
38
2
100% credibility
Found Mar 15, 2026 at 38 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
TypeScript
AI Summary

lat.md creates a connected web of markdown notes about a codebase, linking high-level concepts to code with validation and search tools for humans and AI agents.

How It Works

1
💡 Discover lat.md

You learn about a friendly way to capture your project's big ideas and decisions in simple notes that everyone can follow.

2
📥 Bring it into your project

You add lat.md to your work with one easy step, and it sets up a cozy spot for your notes.

3
✨ Start writing your notes

You jot down key thoughts about your project's design, logic, and plans in plain, readable writing.

4
🔗 Connect the ideas

You link related thoughts together so you can hop from one concept to another effortlessly.

5
📝 Link notes to your work

You add gentle reminders in your files that point back to the right notes, keeping everything tied together.

6
✅ Make sure it all matches

You give it a quick once-over to confirm your notes and work stay perfectly in step.

7
🔍 Ask and find instantly

You describe what you're looking for in everyday words, and it shows you the perfect notes right away.

🎉 Your project comes alive

Now you and your smart helpers easily understand the full picture, making work faster and smarter.

Sign up to see the full architecture

6 more

Sign Up Free

Star Growth

See how this repo grew from 38 to 38 stars Sign Up Free
Repurpose This Repo

Repurpose is a Pro feature

Generate ready-to-use prompts for X threads, LinkedIn posts, blog posts, YouTube scripts, and more -- with full repo context baked in.

Unlock Repurpose
AI-Generated Review

What is lat.md?

lat.md builds a knowledge graph for your codebase using interconnected markdown files in a lat.md/ directory. Wiki links like [[cli#search]] connect concepts across files, while code comments (// @lat: [[section]]) tie implementation back to docs. A TypeScript CLI handles init, validation (lat check), lookup (lat locate/refs), semantic search via OpenAI embeddings, and prompt expansion for agents.

Why is it gaining traction?

It scales beyond monolithic AGENTS.md files, keeping high-level decisions and business logic discoverable as codebases grow—agents hallucinate less with reliable context. Semantic search shines for natural queries like "how do we auth?", and GitHub integrations (agent github claude hooks, agent github copilot, agent github action) make it plug-and-play for AI workflows. No more drift: lat check catches broken links or missing code refs instantly.

Who should use this?

Teams using Claude Code, Cursor, or GitHub Copilot who maintain agent-driven repos with complex logic (auth flows, pipelines). Ideal for backend devs documenting test specs or architecture where flat READMEs fail, or anyone tired of agents ignoring scattered docs.

Verdict

Grab it if you're building agent github repos—early wins on sync and search outweigh rough edges. 38 stars and 1.0% credibility score signal prototype maturity; docs are solid but expect iteration as adoption grows. (187 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.