AI Agent
One agent that knows your factory.
Digel has one AI agent. It runs in the chat sidebar, in @digel mentions inside work order comments, in background triage investigations, and in report generation. Always the same agent, with the same knowledge of your plant.
What the agent knows
Before answering, the agent reads your context graph: asset hierarchy, live sensor data, uploaded documents, maintenance history, glossary, and notes. Every answer includes citations pointing to the specific asset, measurement, or document it used.
Where the agent lives
The chat sidebar is on every page. For longer investigations, use the full-page chat with persistent threads.
Type @digel in any work order comment and the agent answers with the full context of that issue and its linked asset.
In the background, the agent runs triage investigations and writes up what it finds. You can also ask it to generate reports from chat.
Capabilities
- Live telemetry queries and charts
- Document search with inline citations
- Work order creation and status updates
- Root cause analysis across data sources
- Python code execution in a sandboxed environment
- Interactive Streamlit dashboards
- Deep research sub-agents for complex investigations