Show stats display MCP tool
Render KPI summary cards, not charts. Each stat requires `key`, `label`, and `value`, and can also include a strict `format` object, a `diff` object like {"value":12.4}, and a `sparkline` object like {"data":[31,34,36,39,42]}.
- MCP tool name:
Show_Stats_Display - Returns: Show stats display MCP response
- Annotations: Read only, not destructive, idempotent, closed world
- Surfaces: GovTribe AI
Examples
Use stats displays for compact KPI cards. Use Show_Chart when the user needs trends, grouped comparisons, or leaderboards.
Show a capture pipeline KPI summary
{
"tool": "Show_Stats_Display",
"arguments": {
"id": "capture-pipeline-summary",
"title": "Capture pipeline summary",
"description": "Current qualified pipeline for the selected market.",
"locale": "en-US",
"stats": [
{
"key": "pipeline-value",
"label": "Pipeline value",
"value": 42500000,
"format": {
"kind": "currency",
"currency": "USD",
"compact": true
},
"diff": {
"value": 12.4,
"decimals": 1,
"label": "vs last quarter"
}
},
{
"key": "active-pursuits",
"label": "Active pursuits",
"value": 18,
"format": {
"kind": "number",
"decimals": 0
}
},
{
"key": "avg-pwin",
"label": "Average PWIN",
"value": 0.37,
"format": {
"kind": "percent",
"basis": "fraction",
"decimals": 0
}
}
]
}
}Payload patterns
Each stat needs a stable key, visible label, and raw value. Use strict format, diff, and sparkline objects when needed.
Show KPI cards with sparklines
{
"tool": "Show_Stats_Display",
"arguments": {
"id": "market-kpis",
"stats": [
{
"key": "new-opportunities",
"label": "New opportunities",
"value": 31,
"format": {
"kind": "number",
"decimals": 0
},
"sparkline": {
"data": [
21,
24,
27,
31
],
"color": "#125475"
}
}
]
}
}Related articles
- Show stats display MCP response: Review the response shape returned by this tool.
- GovTribe AI: Review the GovTribe AI surface where this tool can be called.
Show search results
Preview search results with rich, visual metadata in an interactive viewer.
Submit teaming match feedback
Submit sentiment + optional text feedback about a teaming match. Reachable for pending and accepted matches only (declined, withdrawn, and expired matches are soft-deleted from agent retrieval). Either participant can submit feedback. The agent should only do so once the user has finished engaging with the match. Sentiment is a boolean (true=positive, false=negative); text is optional, max 2000 chars. Feedback is recorded per-account against the match record.