> ## Documentation Index
> Fetch the complete documentation index at: https://docs.orbitrage.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# Ask Analytics

> In-app assistant that answers natural-language questions about your telemetry — and creates alerts and charts.

Ask Analytics is a chat assistant in the dashboard (bottom-right). Ask anything about your usage; it runs real, org-scoped queries and renders the result as a chart you can pin.

```
"how many calls today?"
"what did we spend this week, by model?"
"which workflow has the highest error rate?"
"alert me when cost goes over $50/day"
```

## How it works

* An LLM with a fixed toolbox of analytics functions.
* It never sees your `org_id` — the server pins it from your session before any tool runs, so the model can only query **your** data.
* Results are capped; secrets are never returned.
* Runs on a fast routed model (DeepSeek-V4-Flash by default) through Orbitrage's own gateway, so its calls show up in your analytics too.

## What it can do

These tools act on your own organization only.

<Tabs>
  <Tab title="Read (anyone)">
    | Tool                                                               | Answers                                                                   |
    | ------------------------------------------------------------------ | ------------------------------------------------------------------------- |
    | `count_calls`                                                      | How many calls in a range, filtered by model/provider/workflow/user/tool. |
    | `sum_cost`                                                         | Total spend, baseline, and savings.                                       |
    | `time_series`                                                      | A metric over time (calls / cost / errors / tokens) for plotting.         |
    | `tool_call_volume`                                                 | How often function-tools were invoked.                                    |
    | `list_models` / `list_end_users` / `list_workflows` / `list_tools` | Leaderboards by traffic or spend.                                         |
    | `detect_anomalies`                                                 | MAD-based anomaly detection on a daily series.                            |
    | `node_drift`                                                       | Per-step drift (tokens, latency, refusal rate) within a workflow.         |
    | `list_chatbot_sessions`                                            | What's been asked recently (UI + MCP), which tools fired, any errors.     |
  </Tab>

  <Tab title="Act (org-scoped)">
    | Tool                           | Does                                                                   |
    | ------------------------------ | ---------------------------------------------------------------------- |
    | `create_alert`                 | Create a webhook/Slack metric alert (returns the signing secret once). |
    | `list_alerts` / `delete_alert` | Manage your alerts.                                                    |
    | `pin_chart`                    | Save a chart to the **Customise** page; it re-runs live.               |
    | `send_to_slack`                | Post a one-off message to your org's Slack channel.                    |
  </Tab>
</Tabs>

## Pinned charts & history

* **Pin** any chart to the **Customise** page, where it re-renders live by replaying the query. Toggle bar / line / donut / table.
* **History** is kept per org in your browser. Switching orgs keeps conversations separate; "New chat" clears just that org's history.

## Everywhere, not just the dashboard

The same toolbox is available from:

* [Slack](/platform/slack) — `@Orbitrage` or `/orbitrage`, with chart images.
* The [MCP server](/platform/mcp) — Claude Desktop, VS Code, or Cursor.

Every turn (UI or MCP) is logged for your own audit via `list_chatbot_sessions`.