Kept AI
Back

Reporting & Automation

Specify a custom visibility dashboard

Turn a dashboard request into a data-backed layout and implementation handoff.

Connect your Kept MCP
Kept AIAI assistantYour workspace
Open Kept AI

Command

Use list_projects({}) to confirm my supplied project host is visible. Use my chosen days value of 7, 30, or 90 for every time-windowed call. Read get_overview({host, days}), get_agent_analytics({host, section: "crawl", days}), and get_agent_analytics({host, section: "conversions", days}). Translate my dashboard questions into available fields, recording units and unknowns. Propose KPI cards, trend tables, source annotations, and access controls. Include a small sample dataset from returned values and an implementation outline for my chosen workspace. Do not invent metrics or execute generated scripts. Hosting, scheduled refreshes, and Notion or other workspace writes require my independently configured connections. Return evidence links, coverage limits, and explicit unknowns; never infer unsupported scores.

What you need

  • A visible host, supported window, dashboard questions, and a chosen display workspace.
  • Independent hosting, scheduling, and workspace connections for deployment or refresh.
Read the setup guide

Map questions to available fields

Define the dashboard around questions the returned data can actually answer. Record field definitions, units, window choices, and missing values before proposing visual components, ensuring that every card has a traceable source and unsupported commercial metrics remain explicitly unavailable.

Hand off a workable dashboard specification

Provide a layout, example values, refresh requirements, and an implementation outline for the chosen workspace. Document access boundaries and external connection needs so an implementer can build the dashboard without mistaking a generated specification for a hosted or automatically refreshed service.