AICoevolution Semantic Telemetry — Standalone Script
Measure conversational dynamics in real-time.
This open-source script demonstrates the core capabilities of the AICoevolution SDK for measuring semantic dynamics in human-AI conversations.
📦 Get the code: github.com/AICoevolution/paper03-orbital-mechanics/open_source
What It Measures
| Metric | Description | Typical Range |
|---|---|---|
| SGI (Orbital Radius) | Balance between query responsiveness and context grounding | 0.5 – 1.5 |
| Velocity | Rate of semantic movement per turn | 0 – 180° |
| Context Phase | Topic coherence state | stable / protostar / split |
| Context Mass | Accumulated turns in current topic | 0 – N |
| Attractor Count | Competing topic centers | 1+ |
Quick Start
1. Install Dependencies
bash
pip install requests2. Get an API Key
Generate a free key at aicoevolution.com/profile → API Keys
3. Run
bash
# Cloud API
python semantic_telemetry.py --api-key YOUR_API_KEY
# Custom turn count
python semantic_telemetry.py --api-key YOUR_API_KEY --turns 20Hosted AI replies (free trial)
During the free trial window, hosted AI is enabled for free keys, and the script can ask the platform to generate the assistant reply for you.
This mode:
- Uses your Telemetry API key for authentication
- Returns an assistant reply plus telemetry computed on the service
- Does not expose any backend URLs or provider keys in the script
Usage (example):
bash
python semantic_telemetry.py --api-key YOUR_API_KEY --hosted-aiIf hosted AI is disabled on your environment, the script will still work in “manual” mode (you provide both sides of the conversation).
Example Output
╔═══════════════════════════════════════════════════════════════════════════════╗
║ SEMANTIC TELEMETRY - AICoevolution SDK ║
╚═══════════════════════════════════════════════════════════════════════════════╝
──────────────────────────────────────────────────
Turn 5/10
──────────────────────────────────────────────────
[YOU]: I'm trying to understand how meaning emerges in conversation
[SDK] <- OK | SGI=0.952, Velocity=34.2°
[AI]: Meaning emerges through the dynamic interplay between what's said
and the accumulated context...
[SDK] <- OK | SGI=0.891, Velocity=28.7°
────────────────────────────── SEMANTIC METRICS ──────────────────────────────
SGI (Orbital Radius): 0.891
Velocity (degrees): 28.7°
Context ID: ctx_1
Context State: stable
Attractor Count: 1
Active Context Mass: 5 turns
──────────────────────────────────────────────────────────────────────Understanding the Metrics
Coherence Region
Productive conversations tend to occupy:
- SGI: 0.7 – 1.3 (balanced orbit)
- Velocity: 15° – 60° (productive movement)
Context Phases
| Phase | Icon | Description |
|---|---|---|
| stable | 🟢 | Anchored to current topic |
| protostar | 🟠 | New topic forming |
| split | 🔴 | Topic changed |
Orbital Energy
E_orb = SGI × Velocity- Higher → more dynamic, potentially unstable
- Lower → more grounded, potentially stagnant
Under Development
Future releases will include:
- Domain Distribution — Cognitive / Somatic / Emotional / Volitional
- Symbolic Depth (S64) — Transformation path detection
- Mass Contribution — Who is steering the conversation?
Learn More
| Resource | Link |
|---|---|
| Paper 03 | Zenodo 10.5281/zenodo.18347569 |
| Website | aicoevolution.com |
| Full SDK Docs | docs.aicoevolution.com |
License
MIT — use freely for research and commercial applications.
Citation
bibtex
@article{jimenez2026orbital,
title = {Semantic Orbital Mechanics: Measuring and Guiding AI Conversation Dynamics},
author = {Jimenez Sanchez, Juan Jacobo},
journal = {Zenodo},
doi = {10.5281/zenodo.18347569},
year = {2026}
}