# plugins

VAOS™ Plugins

Extend the runtime — perception, reasoning, action. Install with one command. Build your own with the typed plugin spec.

Submit a plugin → Plugin spec →
$ vaos plugin install ppe-detector

# perception

P

ppe-detector

v1.4.0 · 2.1k installs

Detects helmets, vests, gloves, and other PPE. Adds attribute fields to person objects.

F

forklift-tracker

v0.7.2 · 540 installs

Multi-camera forklift tracking with heading and speed estimation.

L

lp-reader

v2.0.1 · 980 installs

License plate detection + recognition. Local-only, GDPR-friendly.

A

anomaly-vision

v0.5.0 · 320 installs

Unsupervised anomaly detection for industrial inspection.

# reasoning

S

scene-lm

v1.0.0 · 1.6k installs

Lightweight local LM for scene summarisation and context reasoning.

Z

zone-policy

v0.9.0 · 720 installs

Define zones and per-zone policies (allowed objects, restricted areas, time windows).

R

retro-memory

v0.3.0 · 210 installs

Long-horizon memory plugin. Indexes scenes for natural-language retrieval.

# action

M

mqtt-publisher

v1.2.0 · 3.0k installs

Publish scenes and actions to MQTT topics. Standard for IIoT integration.

S

slack-notify

v0.6.0 · 1.2k installs

Push notifications and rich cards to Slack channels.

R

ros2-bridge

v0.8.0 · 480 installs

Publish VAOS scenes and actions as ROS 2 topics.

P

plc-bridge

v0.4.0 · 140 installs

Connect VAOS actions to industrial PLCs (Modbus, OPC UA).

Build your own

Plugins are typed contracts. Implement the layer's interface, register, ship.

my_plugin.pypython
from vaos.plugin import PerceptionPlugin

class MyDetector(PerceptionPlugin):
    name = "my-detector"
    version = "0.1.0"

    async def on_frame(self, frame, scene):
        # add objects, attributes, or events to the scene
        scene.objects.append({
            "type": "my_object",
            "confidence": 0.92
        })
        return scene
Submit your plugin via PR to github.com/vaos-online/plugins. Reviewed within 7 days. Featured plugins appear on the marketplace.