/install fde-industrial-skill
FDE Industrial AI Deployment Skill
Open Source: https://github.com/jaccen/FDE-Industrial-Skill
Overview
Full-spectrum support for FDEs deploying AI & big data on industrial production lines — from scenario diagnosis to scaled deployment.
Core Workflow
Scenario Diagnosis -> Data Governance -> Solution Design -> POC -> Scale-up -> Feedback Loop
Step 1: Scenario Diagnosis
- Read references/fde-role-model.md for FDE capability framework.
- Apply "Pain-Data-Impact" triage: Pain (business pain), Data (sufficiency), Impact (quantifiable ROI).
- Classify into 5 core categories — references/industrial-ai-scenarios.md.
Step 2: Data Governance & Integration
- Map data sources: OT (SCADA/PLC/sensors), IT (MES/ERP/PLM), ET (engineering docs).
- Palantir-style Ontology: Objects, Links, Actions.
- Data quality gaps: missing values, timestamp misalignment, label scarcity.
- Pipeline: edge collection -> ETL -> feature store.
Key: Start from business decisions, not data tables.
Step 3: Solution Design
- Visual inspection: CNN/ViT + edge GPU boxes
- Predictive maintenance: LSTM/Transformer + physics-informed features; 7-14 day window
- Process optimization: RL/Bayesian + digital twin; single process first
- Energy efficiency: regression + control optimization; baseline first
- Supply chain: graph model + demand forecast + ERP integration
Step 4: POC Deployment (Zero Week)
Day 1-3: data audit + interviews; Day 4-7: baseline model + quick wins; Week 2-4: training + integration; Week 4-6: A/B test + operator training.
Critical: Deliver measurable quick win within 2 weeks.
Step 5: Scale-up & Feedback
Measure ROI, generalize single -> multi -> factory-wide, FDE+FDR feedback loop.
ROI Framework
| Metric | Typical Range |
|---|---|
| Defect detection improvement | 80-95% reduction |
| Unplanned downtime reduction | 30-60% reduction |
| Yield improvement | 2-8% increase |
| Energy savings | 5-15% reduction |
| ROI payback period | 6-18 months |
Reference Guide
| Need | Reference |
|---|---|
| FDE role & skills | fde-role-model.md |
| Scenario & algorithm | industrial-ai-scenarios.md |
| Deployment methodology | landing-methodology.md |
| Case studies | case-studies.md |
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install fde-industrial-skill - After installation, invoke the skill by name or use
/fde-industrial-skill - Provide required inputs per the skill's parameter spec and get structured output
What is Clawhub Publish?
FDE skill for industrial AI deployment: scenario diagnosis, data governance, solution design, POC-to-scale methodology, ROI quantification. Covers predictive... It is an AI Agent Skill for Claude Code / OpenClaw, with 147 downloads so far.
How do I install Clawhub Publish?
Run "/install fde-industrial-skill" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is Clawhub Publish free?
Yes, Clawhub Publish is completely free, licensed under MIT-0. You can download, install and use it at no cost.
Which platforms does Clawhub Publish support?
Clawhub Publish is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).
Who created Clawhub Publish?
It is built and maintained by jaccen (@jaccen); the current version is v2.0.1.