Skip to content

GroundMesh

GroundMesh is a self-healing organizational memory layer that gives every human and AI agent verified, task-specific context before it acts.

Code Video Cerebral Valley

Core trick

Agents tag each stored fact with source and freshness, and an action guard checks proposed actions against verified context before they run.

Tags

agentssecurity

Description

GroundMesh is a self-healing organizational memory layer that gives every human and AI agent verified, task-specific context before it acts. Multiple agents extract company knowledge, detect stale or conflicting information, verify authoritative facts, and generate Context Packs. An Action Guard then allows, warns, blocks, or escalates proposed actions, while approved outcomes are written back into shared memory.
shivanshgupta365/GroundMesh-AI★ 00 forksTypeScriptApache-2.0Created Jul 11, 2026 · last push Jul 11, 2026

README excerpt

# GroundMesh AI

> **Verified context for every human and agent.**

GroundMesh is a self-healing organizational context layer for agentic companies. It turns fragmented company knowledge into structured, source-backed memory, detects stale or conflicting information, builds task-specific Context Packs, and blocks unsafe agent actions before they reach customers or internal systems.

Built for the **Cerebral Valley × Google DeepMind Bangalore Hackathon**.

**Track 2 — Autonomous Orchestration with Managed Agents**

---


## Demo Video

<p align="center">
  <a href="https://www.youtube.com/watch?v=5btGUtNa9Ks">
    <img
      src="https://img.youtube.com/vi/5btGUtNa9Ks/maxresdefault.jpg"
      alt="GroundMesh Demo Video"
      width="900"
    />
  </a>
</p>

<p align="center">
  <strong>▶ Watch the GroundMesh demo</strong>
</p>

> The preview above is displayed directly in the README and opens the full demo when clicked.

---

## Table of Contents

- [Demo Video](#demo-video)
- [The Problem](#the-problem)
- [What GroundMesh Does](#what-groundmesh-does)
- [Core Idea](#core-idea)
- [Key Concepts](#key-concepts)
- [Agent Team](#agent-team)
- [System Architecture](#system-architecture)
- [End-to-End Workflow](#end-to-end-workflow)
- [Product Experience](#product-experience)
- [Demo Scenario](#demo-scenario)
- [Technology Stack](#technology-stack)
- [Repository Structure](#repository-structure)
- [Getting Started](#getting-started)
- [Environment Variables](#environment-variables)
-

Team