WebSocket Protocol: cc-mcs ↔ EchelonCapture

Full HTML reader

Read the full artifact

Extracted abstract or opening context

Currently EchelonCapture **sends** sensor data to cc-mcs via HTTP POST. Now we need **bidirectional communication** so EchelonCapture can **receive** visualization data. **Query params:** - `device_id`: Which device this is (left, right, watch) - Optional: `subscribe`: Comma-separated list of channels (latent,trajectory,pattern,audio) ### Authentication None required for now (same network assumption as existing HTTP) | Message Type | Frequency | Bandwidth | Priority | |--------------|-----------|-----------|----------| | `latent_state` | 50 Hz | ~2 KB/s | 🔴 Critical | | `trajectory` | 20 Hz | ~3 KB/s | 🟡 High | | `audio` | 30 Hz | ~1 KB/s | 🟡 High | | `pattern` | 1-4 Hz | ~0.5 KB/s | 🟢 Medium | | `section` | 0.01-0.05 Hz | ~0.01 KB/s | 🟢 Medium | | `ambient_aura` | 5 Hz | ~0.2 KB/s | ⚪ Low | EchelonCapture should: 1. Show connection lost indicator 2. Attempt reconnect every 2 seconds 3. Continue sensor streaming (HTTP still works) 4. Resume visualization when reconnected

Promotion decision

What has to happen next

Attach run IDs, datasets, metrics, and reproduction commands.

Why this is not always a full paper yet

Corpus pages are public-safe readers for discovered workspace artifacts. They are not automatically final papers. A corpus item becomes a polished paper only after the editable source, evidence checkpoints, references, figures, render path, and release status are attached through the paper schema.