Platform Dashboard — User Guide
Audience: Everyone who lands on Dagen after sign-in and wants to understand Platform Overview, summary cards, and Recent Tasks.
Overview
After you authenticate, the home experience is usually the Platform Overview dashboard (also reachable from Overview / Dashboard in the main navigation). It answers at a glance:
- How many database connections, integrations, agents, and data pipelines you have
- Whether anything looks unhealthy (failed pipelines, disabled integrations, connection issues)
- What just ran (ingestion, insights, test data, pipeline jobs) and where to go for details
The dashboard is the fastest way to jump into management screens without hunting the sidebar.
Main sections
Title area
- Platform Overview heading with a short subtitle
- Sets context: this is your workspace’s operational snapshot, not a deep drill-down
Summary cards (top row)
Four cards summarize major asset classes. Each card loads independently (its own spinner or error state).
| Card | What you see | Primary action |
|---|---|---|
| DB Connections | Count of configured connections; status such as healthy / unhealthy connections or “Status OK” | Manage → Database Connections (/db-connections) |
| Integrations | Count of integrations; note if any are disabled | View All → Integrations (/integrations) |
| Agents | Total agents, often split into Builder-defined vs built-in (hardcoded) | Build / Manage → Agent Builder; tune behaviour in Agent Intelligence & Skills (/agent-intelligence) |
| Data Pipelines | Total pipelines; may show breakdown by type (e.g. dbt, Dataform); flags failing or running pipelines | View All → Building Pipelines (/pipelines) |
Use cases
- On-call check: Open home first to see if pipelines or connections are red.
- Onboarding: Confirm connections and repos exist before asking agents to build.
- Capacity: See how many custom agents and tools the team has created.
Recent Tasks
Below the cards, Recent Tasks lists platform activity (not only chat).
| Element | Purpose |
|---|---|
| Refresh | Reloads the task list (shows loading while fetching) |
| Task rows | Each row: icon by task type (ingestion, insights, test_data, pipeline, etc.), name, type, relative time (“Just now”, “5 min ago”), optional source snippet |
| Status | Chip: completed, running, failed, etc.; running tasks may show a progress indicator |
| View All Tasks | Opens Job History (/job-history) for the full log |
Empty state: If nothing has run yet, you’ll see a message instead of an empty table—normal for new workspaces.
Use cases
- Did my ingestion finish? Scan the list without opening the ingestion UI.
- Find a failed pipeline run and jump to pipelines or job history from there.
- Correlate chat-driven jobs with background tasks.
How this fits the rest of the product
| Goal | From the dashboard |
|---|---|
| Fix a connection | DB Connections → Manage |
| Tune agents / skills | Agents → Build / Manage → also Agent Intelligence |
| Debug a transformation | Data Pipelines → View All; use Fix with Agent on failures (Building Pipelines) |
| Full job audit | Recent Tasks → View All Tasks |
| Ask anything in natural language | Open AI Chat from the nav |
Technical note (for integrators)
The dashboard aggregates data from standard APIs (connections, integrations, agents, pipeline summary, recent tasks). If a section errors, others may still load—retry or check network and permissions.
Related
- Introduction — Platform vision and agent hierarchy
- Administration — Job history, runtimes, usage
- Data Ingestion — Sync monitoring in depth