Skip to content

neha037/mesh

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

9 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Mesh

A localized, private Personal Growth Engine that maps structured technical knowledge and creative pursuits into a unified, interactive knowledge graph — with algorithmic serendipity to combat intellectual stagnation.

Why Mesh?

Every existing PKM tool is passive. They rely on you to manually create connections and review old material. Mesh is an active cognitive partner that:

  • Visualizes your knowledge as an interactive graph
  • Automatically extracts concepts and builds connections
  • Detects knowledge gaps and over-saturated clusters
  • Injects serendipitous cross-domain discoveries
  • Uses spaced repetition to fight knowledge decay

All running locally on your machine — zero cloud costs, total data sovereignty.

Architecture

┌──────────────────┐
│  React Frontend  │ ← Cytoscape.js graph, search, filters, review cards
│    (Port 3000)   │
└────────┬─────────┘
         │ REST/JSON
┌────────┴─────────┐
│   Go API Server  │ ← chi router, ingestion, graph queries, search
│    (Port 8080)   │
└────────┬─────────┘
         │
┌────────┴─────────┐
│  Go Workers      │ ← scraping, NLP tagging, embedding, edge building
└──┬─────┬─────┬───┘
   │     │     │
   ▼     ▼     ▼
 PG16  MinIO  Ollama

Tech Stack

Layer Technology Purpose
Backend Go + chi REST API and background workers
Database PostgreSQL 16 + pgvector Graph storage, vector search, job queue
Object Store MinIO Image and file storage (S3-compatible)
AI/NLP Ollama (local) Tag extraction, embeddings, summaries
Frontend React + TypeScript + Cytoscape.js Interactive graph visualization
Orchestration Docker Compose Service topology and lifecycle

Quick Start

Prerequisites

  • Docker and Docker Compose
  • Git

Run

# Clone the repository
git clone https://github.com/neha037/mesh.git
cd mesh

# Configure environment
cp .env.example .env
# Edit .env and set PG_PASSWORD and MINIO_PASSWORD

# Start core services (PostgreSQL, MinIO, API, Worker, Web)
cd deploy && docker compose up -d

# Run database migrations
make migrate-up

# (Optional) Start Ollama for AI features
docker compose --profile ai up -d ollama
make pull-models

The API will be available at http://localhost:8080. The web dashboard is planned for Phase 4.

Verify

# Save a test page
curl -X POST http://localhost:8080/api/v1/ingest/raw \
  -H "Content-Type: application/json" \
  -d '{"url": "https://example.com", "title": "Test Page", "content": "Hello world"}'

Development

See the Developer's Guide for detailed setup, workflows, and conventions.

# Run with hot-reload (development mode)
cd deploy && docker compose -f docker-compose.yml -f docker-compose.dev.yml up

# Run tests
make test

# Run linter
make lint

Project Status

Current Phase: Phase 3 — Graph Traversal & Query API ("The Memory")
Last Verified: April 8, 2026 — Phase 2 complete, committed, and production-ready

Phase Name Status
1 Foundation & Ingestion — "The Senses" ✅ Complete (API, ingestion, scraper with robots.txt, job queue, worker pool, circuit breaker)
2 Processing & Intelligence — "The Brain" Complete & Committed (Ollama with circuit breaker, AI tagging, embeddings, semantic similarity, auto edge building, fallback NLP, 36 tests)
3 Graph Traversal & Query API — "The Memory" ⏳ Ready to start (BFS, full-text search, semantic search, hybrid search, pagination)
4 Frontend Visualization — "The Eyes" Not started
5 Multi-Modal & Journaling — "The Human Element" Not started
6 Anti-Echo Chamber Engine — "Discovery" Not started
7 Spaced Repetition & Semantic Depth — "The Slow Burn" Not started

Documentation

Design Constraints

  • Zero cloud costs — all compute and storage runs locally
  • Single developer — 6-8 hours/week sustainable pace
  • Absolute data sovereignty — no data leaves the machine
  • Standard hardware — runs on 16 GB RAM workstation
  • Ephemeral compute — safe to shut down at any time

About

No description, website, or topics provided.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors