proffesor-for-testing

proffesor-for-testing / nagual-qe

Public

Self-learning knowledge system for quality engineering. Rust, SQLite + optional PostgreSQL, ONNX embeddings, hybrid retrieval, Bayesian quality scoring.

agentic-qe ai-agents knowledge-management quality-engineering reasoning-bank
11
1
100% credibility
Found Apr 23, 2026 at 11 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
Rust
AI Summary

Nagual-QE is a local-first, privacy-respecting tool that stores quality engineering patterns, learns from outcomes using Bayesian scoring, and retrieves them via hybrid search for engineers and AI agents.

How It Works

1
📰 Discover a smart testing helper

You hear about a simple tool that remembers your best fixes for flaky tests and bugs, helping you work faster without forgetting what works.

2
💻 Get it set up quickly

Follow easy steps to install it on your computer, and it creates a private notebook just for your testing knowledge.

3
✨ Save your first trick

Type in a problem like 'flaky async test' and your solution, add a testing tag, and it stores it safely with a smart score.

4
🔍 Find fixes instantly

When a similar issue pops up, search naturally and get your proven solutions ranked by what worked best before.

5
📈 Help it learn from results

After trying a fix, tell it if it succeeded or failed, so it updates scores and gets better at suggesting winners.

🚀 Your testing gets supercharged

Over time, it builds a personalized collection of your top patterns, dashboard shows progress, and you solve issues faster every day.

Sign up to see the full architecture

4 more

Sign Up Free

Star Growth

See how this repo grew from 11 to 11 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 nagual-qe?

Nagual-qe is a Rust-built self-learning knowledge system for quality engineering, storing problem-solution patterns with tags and embeddings while learning their effectiveness through Bayesian quality scoring from real outcomes. It retrieves relevant knowledge via hybrid search combining full-text, cosine similarity on ONNX embeddings, and graph boosts, all in a local-first setup with SQLite or optional PostgreSQL. Users get a privacy-focused memory layer that predicts pattern success with calibrated confidence, redacting PII before any cloud sync.

Why is it gaining traction?

In a sea of generic self learning ai agents github projects, nagual-qe stands out with QE-specific seed patterns for instant value, plus self-learning knowledge graph features like automatic duplicate consolidation and outcome-driven Bayesian updates over simple rewards. Developers hook it into AI workflows via CLI commands like `nagual knowledge store` and `learn record`, or its HTTP/WebSocket dashboard and Unix socket API, enabling self directed learning metacognitive knowledge includes without vendor lock-in.

Who should use this?

Quality engineers debugging flaky tests, test automation specialists tracking patterns across CI runs, or engineering teams building self learning agent github tools for hybrid retrieval in agentic QE fleets. It's ideal for those needing stats self learning github capabilities with embeddings and bayesian scoring to surface working solutions fast.

Verdict

Pre-1.0 with just 11 stars and 1.0% credibility score, it's early but actively developed with solid docs, benchmarks, and Docker support—worth prototyping for Rust-savvy QE niches if you seed your own patterns. Skip for production until migrations stabilize.

(198 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.