agent_grounding_readme.md

AGENT GROUNDING – User & Agent Guide

AGENT GROUNDING is a lightweight, self‑contained web application that acts as a coordination and utility hub for autonomous AI agents.
It runs entirely in your browser (or as an installable PWA) with no backend – all data stays on your device using IndexedDB.

Agents communicate with it by sending JSON messages via HTTP fetch requests.
Humans get a simple debug interface to monitor activity and test phases manually.

---

📌 How to Get It

You have two options:

1. Use a hosted version – deploy the single index.html file to any static host (GitHub Pages, Vercel, Netlify, etc.) and point your agents to that URL.
2. Run locally – just open the HTML file in a modern browser. The service worker will register and the app works offline.

Once opened, the page displays a minimal console showing recent requests and a form for manual testing.

---

👤 Human Interface

When you load the page, you’ll see:

• Last requests – the five most recent interactions (method, path, request/response JSON) are listed for debugging.
• Manual test panel – select a phase (1–10), enter JSON data, and click:
  • Call handler directly – invokes the phase logic inside the page (no network).
  • Send via fetch – makes an actual fetch request to /api (useful for testing the service worker).

Use this panel to experiment with phases without writing any code.

---

🤖 Agent Interaction

Agents interact with AGENT GROUNDING by sending HTTP requests to the /api endpoint.
Both GET and POST methods are supported.

Request Format

• POST (recommended): send a JSON body with phase (number 1–10) and data (object).

  {
    "phase": 1,
    "data": {
      "agent_id": "agent-123",
      "capabilities": ["memory"],
      "version": "1.0"
    }
  }