Luispitik

Luispitik / sinapsis

Public

Continuous learning system for Claude Code. Observes your patterns, creates instincts, evolves skills. Based on ECC v2.1.

18
2
89% credibility
Found Mar 18, 2026 at 18 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
Shell
AI Summary

Sinapsis is a continuous learning system for AI coding assistants that captures session patterns and evolves them into reusable instincts and skills.

How It Works

1
📖 Discover Sinapsis

You hear about Sinapsis, a helper that makes your AI coding buddy learn from every chat session to get smarter over time.

2
📥 Get and run the installer

Download the simple setup file and run it to create special folders where your AI will store memories.

3
🧠 Add your personal touch

Fill in a bit about yourself so the system knows who you are and tailors learning to your style.

4
👁️ Turn on watching

Optionally tell your AI to watch what happens in sessions, or start using special chat commands right away.

5
💬 Chat and work with AI

Use your AI as usual for coding tasks, and it quietly notes patterns and successes from each interaction.

6
✨ Watch instincts grow

Check commands to see new learned behaviors form, like shortcuts for your common fixes or workflows.

🎉 Smarter AI companion

Over time, your AI remembers your ways, repeats wins, and helps you code faster with personalized smarts.

Sign up to see the full architecture

5 more

Sign Up Free

Star Growth

See how this repo grew from 18 to 18 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 sinapsis?

Sinapsis is a shell-based continuous learning system for Claude Code that observes your tool usage patterns across sessions, logs them securely per Git project, and turns them into evolving "instincts" – reusable skills tailored to your workflow. It solves the problem of repetitive AI interactions by automating pattern detection and skill evolution, much like continuous learning machine learning applied to your coding habits. Run the installer, add hooks to your Claude settings, and use commands like /instinct-status or /evolve to manage it.

Why is it gaining traction?

It stands out by hooking directly into Claude's Pre/PostToolUse events for passive observation, auto-detecting projects via Git remotes, and providing commands to cluster patterns, promote instincts globally, or export them – delivering continuous learning and development without manual logging. Developers dig the project-specific registries and secret scrubbing, turning sessions into a personal continuous learning AI model that feels like a synonym for smarter assistance. With low overhead, it's a fresh take on github continuous integration for AI-driven coding.

Who should use this?

Claude Code power users building in Git repos who want AI that adapts to their stack via /dna or captures fixes with /gotcha. Devs in continuous delivery github actions pipelines tired of reteaching tools, or teams practicing continuous learning examples in machine learning workflows. Ideal for solo hackers evolving skills across projects, not beginners.

Verdict

Try it if you're deep in Claude Code and value continuous thought machine vibes – the 18 stars and 0.9% credibility score signal early-stage maturity with solid docs but no tests yet. Solid seed for personalizing AI, just watch for hook tweaks in diverse setups.

(198 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.