mcpbeat

Fleet Intelligence

microsoft/fleet-intelligence

Monitor robot fleet telemetry via Azure IoT Operations, drift detection, Grafana dashboards, and Fabric analytics

600 tokens
context cost
the whole folder, loaded on every use
1
files
instructions only
0
copies elsewhere
how many repositories repackaged it
100
stars on the repo
on the repository, not the skill itself

Install

one command, takes just this skill from the repository
npx skills add https://github.com/microsoft/physical-ai-toolchain --skill fleet-intelligence

The instruction itself

10 sections, as written by the author

Fleet Intelligence

Monitor and analyze deployed robot fleet telemetry — IoT Operations data collection, Grafana dashboards, drift detection, and Fabric Real-Time Intelligence analytics.

Prerequisites

| Requirement | Purpose |

|-------------|---------|

| Azure IoT Operations | Edge telemetry collection |

| Azure Event Hubs | Cloud telemetry ingestion |

| Grafana | Fleet dashboards and alerting |

| Microsoft Fabric | Real-Time Intelligence KQL analytics |

| az CLI | Azure authentication |

| kubectl | Cluster access for IoT Operations |

Setup Workflow

Deploy fleet intelligence components in order:

Step 1 — Deploy IoT Operations

fleet-intelligence/setup/deploy-iot-operations.sh

Step 2 — Deploy telemetry pipeline

fleet-intelligence/setup/deploy-telemetry-pipeline.sh

Step 3 — Deploy dashboards

fleet-intelligence/setup/deploy-dashboards.sh

Step 4 — Deploy Fabric Real-Time Intelligence

fleet-intelligence/setup/deploy-fabric-rti.sh

Telemetry Schemas

| Schema | Path | Description |

|--------|------|-------------|

| Policy Execution | telemetry/schemas/policy-execution.schema.json | Inference metrics and action outputs |

| Robot Health | telemetry/schemas/robot-health.schema.json | Hardware status and connectivity |

Drift Detection

Drift detection monitors deployed policies for performance degradation. When drift exceeds configured thresholds, retraining triggers initiate automated training pipelines.

| Component | Path | Description |

|-----------|------|-------------|

| Detection | drift/detection/ | Statistical tests against training baselines |

| Alerting | drift/alerting/ | Threshold evaluation and notification routing |

| Triggers | drift/triggers/ | Automated retraining pipeline launching |

Key Files

| File | Description |

|------|-------------|

| fleet-intelligence/README.md | Domain overview |

| fleet-intelligence/telemetry/routing/edge-to-eventhub.yaml | Edge-to-cloud routing config |

| fleet-intelligence/dashboards/grafana/fleet-overview.json | Grafana dashboard definition |

| fleet-intelligence/dashboards/fabric/fleet-kql-queries.kql | Fabric KQL analytics queries |

How to use it

Copy the folder

Take microsoft/fleet-intelligence from the repository into ~/.claude/skills for personal use, or into .claude/skills inside a project.

Check the name does not clash

The agent identifies a skill by the name field in its header. Two skills with the same name cannot sit side by side — one of them will be ignored.