kromahlusenii-ops

kromahlusenii-ops / ham

Public

(HAM) Memory system for AI coding agents. Cut token usage by 50% by scoping context to directories.

ham-pro.vercel.app agent-memory agent-skills ai-coding-assistant claude-code claude-code-skill
22
1
100% credibility
Found Feb 26, 2026 at 16 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
TypeScript
AI Summary

HAM is a system of folder-specific notes that lets AI coding assistants load only relevant project context, reducing usage by up to 50% while providing a local dashboard to track savings.

How It Works

1
📖 Discover HAM

You hear about HAM, a clever way to help your AI coding assistant remember just the right project details without wasting effort.

2
🛒 Bring HAM home

You easily add HAM to your coding tools folder so it's ready to use.

3
✨ Add folder notes

You create short, helpful notes in your project folders to guide your AI on what matters in each spot.

4
🤖 AI works smarter

Your AI buddy now reads only the nearby notes it needs, making everything quicker and lighter.

5
📊 See your savings

You open a friendly web dashboard that shows colorful charts of time, money, and energy you've saved.

🎉 Code greener & faster

Now your projects run smoother with less cost and a smaller footprint on the planet.

Sign up to see the full architecture

4 more

Sign Up Free

Star Growth

See how this repo grew from 16 to 22 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 ham?

HAM builds a hierarchical memory system for AI coding agents like Claude Code, replacing bloated single context files with scoped CLAUDE.md notes per directory to cut token usage by 50%. You get CLI commands like "go ham" for quick setup, "HAM audit" for health checks, and "HAM dashboard" for a local web view of session stats, savings, and context coverage. JavaScript-powered, it parses agent logs to visualize input tokens, costs, and directory trends without external services.

Why is it gaining traction?

Unlike generic prompt libraries, HAM auto-scopes context to your working directory, self-maintains via agent instructions, and proves ROI with a dashboard showing real token/cost reductions—up to 50% on prompts. Devs hook on the "HAM sandwich" fun CLI alias and greentext sustainability angle, plus it works across agents without code changes. The github ham dashboard delivers personalized insights like cache efficiency and routing adherence.

Who should use this?

Claude Code power users in large repos tired of reloading full project context every session. Teams tracking AI API bills for frontend/backend work in src/api or components dirs. Solo devs optimizing token spend before scaling to Cursor or Copilot.

Verdict

Try it—solid README and working dashboard make onboarding fast, despite 15 stars and 1.0% credibility signaling early maturity. No tests visible, but zero-setup CLI delivers immediate 50% wins if context bloat hits you.

(198 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.