> For the complete documentation index, see [llms.txt](https://help.tellius.com/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://help.tellius.com/tellius-6.4/kaiya/kaiya-missions/how-to-create-a-kaiya-mission/the-mission-pane.md).

# The Mission pane

Review and change a Kaiya Mission in the Mission pane: the pane header and the Steps, Preview, Triggers and Schedule tabs.

The Mission pane opens on the right side of the screen, next to the Mission conversation. This page covers the pane header and its four tabs. For the whole flow, from the first request to the first scheduled run, see [How to create a Kaiya Mission](https://help.tellius.com/kaiya/kaiya-missions/how-to-create-a-kaiya-mission).

### **The pane header**

* **Mission name and description:** Kaiya writes the description and keeps it current as the Mission changes.
* **Version:** shows which version you are looking at, such as **V1**. Select it to list the Mission's versions and switch between them. See [Versions](https://help.tellius.com/kaiya/kaiya-missions/how-to-create-a-kaiya-mission#versions).
* **Refresh icon:** re-runs the whole Mission now on fresh data, the same as **Run**, using the published (active) version. Before the run starts, Tellius asks you to confirm in a **Refresh this mission?** dialog. The icon is greyed out until the Mission has a published version, and it is unavailable for about 30 seconds after you select it. You can also ask Kaiya in the conversation, for example "Refresh the Weekly Sales mission". Kaiya runs the published version, never a draft.
* **Open full view (arrows icon):** opens the Mission's own page, with its latest briefing and run history. See [Opening a Mission from the Mission Library or the home page](https://help.tellius.com/kaiya/kaiya-missions/how-to-create-a-kaiya-mission/mission-library-and-mission-page#opening-a-mission-from-the-mission-library-or-the-home-page).
* **Close (x):** closes the pane. The conversation stays open.

If the Mission changes while you have it open, a banner tells you so. Select **View latest** to load the newest version.

The pane has four tabs: **Steps**, **Preview**, **Triggers** and **Schedule**.

{% hint style="info" %}
**When a Mission is shared with you:** the pane is read-only. You can look at the **Steps**, **Preview**, **Triggers** and **Schedule** tabs, but you can't change them. **Schedule** appears only once the Mission is published. The header has no **Version** selector, because you always see the owner's live (active) Mission. The pane updates on its own when the owner publishes a change, so the banner about changes doesn't appear. If your role can create Missions, the header also has a **Remix** button that saves your own copy to change. See [Remix a Mission shared with you](https://help.tellius.com/kaiya/kaiya-missions/how-to-create-a-kaiya-mission/remix-a-mission-shared-with-you).
{% endhint %}

### **Steps tab**

The **Steps** tab lays out the plan Kaiya built from your request, one step at a time, in the order the steps run. The number next to the tab name is the step count. Use the switch at the top right to choose **Outline** or **Diagram**.

**Outline** lists the steps in order inside an **Analysis Steps** card. The number after the card title is the step count. This view suits reading a plan from start to finish.

The card header holds two items:

* **Locked:** shown when you are looking at the active published version. A published version can't be changed. An earlier published version shows **Read-only** instead, so you can see how the Mission used to work. To change the plan, send a message in the conversation, and Kaiya creates a draft for your changes. See [Versions](https://help.tellius.com/kaiya/kaiya-missions/how-to-create-a-kaiya-mission#versions).
* **Copy icon:** copies the whole step list.

Each step row shows:

* **Step number:** the order the step runs in, such as **01**.
* **Type badge:** the kind of step, such as **SQL**, **Python**, **Summary** or **Rich Summary**.
* **Description:** a one-line summary of what the step does.

Select the arrow on a step to expand it. An expanded step shows:

* **WHAT THIS STEP DOES:** a plain-language explanation of the step, in five parts: **Goal**, **What data is gathered**, **Filters applied**, **How it's processed** and **Output**. Read it to check the step's logic without reading any SQL. Use the copy icon beside the heading to copy the explanation.
* **TEMPLATIZED SQL:** the SQL behind the step. Select the arrow to show or hide it, and the copy icon to copy it.
* **Output:** the name of the result the step produces, such as `region_week_scorecard`, and the **Business View** the step reads. Later steps use the output name to read the result.

The sections of an expanded step depend on its type:

| Step type                        | What you see when you expand it                                                                                                                                           |
| -------------------------------- | ------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
| **SQL**                          | **WHAT THIS STEP DOES**, then the SQL: **TEMPLATIZED SQL**, or **SQL INSTRUCTIONS** for a step written as instructions.                                                   |
| **Python**                       | **WHAT THIS STEP DOES**, then **CODE**, the Python code the step runs.                                                                                                    |
| **Input**                        | **WHAT THIS STEP COLLECTS**, the value the step asks for and how, then **CODE**.                                                                                          |
| **Summary** and **Rich Summary** | **SUMMARY PROMPT**: the instructions the step follows to write the briefing.                                                                                              |
| **Export**                       | **STORYLINE**: the storyline and contents of the file the step creates.                                                                                                   |
| **Connected system**             | **REQUEST**, what the step asks the connected system, or **TOOL** and **ARGUMENTS** for a step that reads with a fixed call. The step's access settings show beside them. |

**Edit steps in the pane.** While you work on a draft, you can change a step in place instead of asking in the conversation:

1. Select a step's title to rename it, or select the text of a section, such as **WHAT THIS STEP DOES**, the SQL, the Python code, the summary prompt or the storyline, to change it.
2. Type your change. The step shows an **EDITED** badge and a note that the change is staged and not yet applied. Select **Revert this step** to drop the changes to that one step.
3. A bar at the bottom of the tab shows how many steps you edited, and that the plan still runs as it was. Select **Apply Changes** to apply all your edits together, or **Discard All Changes** to drop them.

Kaiya applies your staged edits as one change, so a plan that needs many small corrections is revised in one go. The edit shows in the conversation as an **\[Inline edit]** note, and Kaiya checks whether any later step needs to change as well. The sections of a connected-system step can't be edited in the pane. Steps can't be edited on a published version, on a Mission shared with you, or while Kaiya is working.

{% hint style="info" %}
**Edit in the pane or ask Kaiya?** Edit in the pane when you know the exact change, such as a filter value in the SQL or a sentence in the summary prompt. Ask Kaiya in the conversation when you'd rather describe the result you want, such as a new measure or a new finding, and let Kaiya work out which steps to change.
{% endhint %}

<figure><img src="https://1702728550-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FC5PhRxmOjkQZJEO6F9Zw%2Fuploads%2FJ1QZDSSGQtGt1RIbbHzz%2Fimage.png?alt=media&amp;token=02506508-4685-4313-aa87-27c48b77b69a" alt="" width="563"><figcaption></figcaption></figure>

**Diagram** lays out the same plan as a map. Each step is a box showing its number, type, description and output. Lines run from each step to the steps that use its result. For example, several SQL steps can all feed one Python step, which then feeds the summary.

Use the **+** and **–** buttons to zoom in and out, and the frame button to fit the whole plan on screen. A small overview map in the corner helps you find your place in a large plan.

{% hint style="info" %}
Switch to **Diagram** when a Mission branches, or when one step draws on two earlier steps. The lines show at a glance which result feeds which step.
{% endhint %}

### **Preview tab**

The **Preview** tab shows what each step produced in a test run, so you can check the result before you publish. Before the Mission has run, it shows how the briefing will look once it runs. For a draft, it has three parts:

* **Planned steps:** every step, with a check mark once it has completed. Expand a step to see what it returned, including its code, its status, its row and column counts, a five-row preview and its run time. Files a step creates, including HTML reports, appear as cards in that step. They open in the Artifact view and stay available after you reload the page.
* **Generated summary:** the full briefing the Mission writes, with the figures it reports.
* **Charts:** the charts the Mission produces, each with its own title.

<figure><img src="https://1702728550-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FC5PhRxmOjkQZJEO6F9Zw%2Fuploads%2F9Tqgs84XNzRJ0XINz2By%2Fimage.png?alt=media&amp;token=e2289961-b8fa-470a-aeaa-d6fca0c38c59" alt="" width="563"><figcaption></figcaption></figure>

Use Preview to make sure a draft gives the right answer while it is still a draft.

For a published version, Preview shows the run that was made when that version was built, and the date it was saved. Runs after that don't appear here. You find them in the conversation and in the Mission's run history. A switch at the top right chooses between two views:

* **Rich Summary:** the visual briefing, with its headline and the cards for the key figures.
* **Report:** **Planned steps**, grouped by iteration, such as **Iteration 1**. Each step is listed as **Step 1**, **Step 2** and so on, with its type and description and a check mark once it completed. Expand a step to see its details.

<figure><img src="https://1702728550-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FC5PhRxmOjkQZJEO6F9Zw%2Fuploads%2FHCUnPTbBOlUNPSmJI9Io%2Fhd-preview-published.png?alt=media&amp;token=deccc37c-6a0e-40a4-bca0-6e44edb9cfff" alt="Preview of a published version, in the Rich Summary view" width="563"><figcaption></figcaption></figure>

**Citations.** Preview offers citations on request. Open them to see where each number in the output came from or how it was derived. Each figure is marked **Verified** or **No source**. Citations are generated only when you request them in Preview, never during scheduled runs.

{% hint style="info" %}
Before you schedule a Mission, ask Kaiya to confirm the numbers, for example *"Check that these figures are correct before I schedule this."* Kaiya runs its own queries to check them and traces where a difference comes from. Where a step looks wrong, Kaiya explains what it found, offers a fix, and can update the steps so the output matches what you asked for.
{% endhint %}

### **Triggers tab**

The **Triggers** tab lets a Mission run when someone asks Kaiya a matching question, not only on a schedule. The **From Kaiya Chat** card holds a list of phrases. Each phrase is a question the Mission can answer. For example:

* *"How did MRR and churn move by region last week?"*
* *"Which region's MRR moved most week over week?"*
* *"Where is churn risk highest this week?"*

<figure><img src="https://1702728550-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FC5PhRxmOjkQZJEO6F9Zw%2Fuploads%2FYeHJle5452WjJekd6Wt2%2Fimage.png?alt=media&amp;token=4233f861-53a3-4c90-984e-2326118eaf9e" alt="" width="563"><figcaption></figcaption></figure>

To add a phrase, type it in the box at the bottom of the card (its hint text starts with **Add a sample query**) and press Enter. Select the **x** beside a phrase to remove it. Changes save straight away, so you don't need to publish them. A phrase that is empty, or already in the list, isn't added. Only the Mission's owner can change the phrases. You can also ask Kaiya in the conversation to change them.

{% hint style="success" %}
**Write several ways of asking.** Kaiya compares a question with each phrase on its own, so add the different ways people ask the same thing, for example *"Why did revenue drop?"*, *"Explain the decline in revenue"* and *"What's causing the revenue dip?"* Kaiya can match a question to a Mission only after the Mission is published, and only for people who can see the Mission.
{% endhint %}

When someone asks a question in Kaiya chat that matches one of the phrases, Kaiya can run this Mission to answer it. The answer names the Mission, the version that ran and how the run was started (Scheduled, Chat, or Run manually). Select the Mission name or the arrow beside it to open that Mission.

The card has an on/off switch at the right.

### **Schedule tab**

The **Schedule** tab sets the Mission to run on its own and decides where each run's briefing goes.

<figure><img src="https://1702728550-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FC5PhRxmOjkQZJEO6F9Zw%2Fuploads%2FQn3himzd9zmBWUH5g4kQ%2Fimage.png?alt=media&amp;token=77d7a0e9-a5ac-42a5-806c-d915bb8dae0f" alt="" width="563"><figcaption></figcaption></figure>

* **Schedule Name:** filled in for you as the Mission name followed by "schedule". Change it to something your readers will recognise, because the name also appears in the **Scheduled Missions** list. See [Scheduled Missions](https://help.tellius.com/kaiya/kaiya-missions/how-to-create-a-kaiya-mission/mission-library-and-mission-page#scheduled-missions).

**Cadence:** sets when the Mission first runs and how often it repeats after that.

* **Start Date:** the date of the first scheduled run. Select the field to open a calendar. Use the month list or the arrows to move between months, select a day, and select **Apply**. Days before today are greyed out. **Cancel** closes the calendar without changing the date.
* **Start Time:** the time of day the Mission runs. Select the field to choose the hour (**01** to **12**), the minutes (in five-minute steps) and **AM** or **PM**, then select **Apply**. Times are shown in your browser's time zone. This includes the times Kaiya gives in the conversation, such as when it confirms a schedule or reports when a run happened.
* **Frequency:** how often the Mission repeats after its first run: **Hourly**, **Daily**, **Weekly**, **Monthly**, **Quarterly** or **Yearly**. **Daily** is selected by default. Choosing a frequency adds no other fields, such as a day of the week, so the start date and start time set when the runs happen. For example, a **Daily** schedule that starts on 11 October at 3:30 AM runs every day at 3:30 AM.

<figure><img src="https://1702728550-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FC5PhRxmOjkQZJEO6F9Zw%2Fuploads%2Fd1noi3e1tTROUYqvsQaf%2Fhd-frequency-and-cc-bcc.png?alt=media&amp;token=61cf0547-c434-4c0b-bde1-d64d7f922fc5" alt="The Frequency list open, with the CC and BCC fields shown under Recipients" width="563"><figcaption></figcaption></figure>

**Delivery:** Set up where each scheduled run's briefing is sent. Turn on a channel with its switch. You can turn on more than one, and each enabled channel gets the same briefing. The counter, such as **1 of 3 enabled**, updates as you turn channels on or off.

* **Email:** turned on to start. The card shows how many people the briefing goes to. **Recipients** starts with your own account email. Type a name or an email address to add more people, and select the **x** on a person to remove them.
  * **Cc/Bcc:** opens a **CC** field and a **BCC** field. People in **CC** get a copy that everyone can see. People in **BCC** get a copy, and the other recipients can't see that they got it.
* **Slack:** Posts this schedule's briefing to Slack. You can choose either to DM you or post in a Slack channel. If your workspace isn't connected to Slack yet, the card says so. When you turn the switch on, the card shows a **Connect** link so you can connect Slack.
* **Microsoft Teams:** Posts this schedule's briefing to Microsoft Teams. You can choose either to DM you or post in a Teams group. When you turn the switch on, two choices appear:
  * **Post to a channel:** selected by default. Pick the channel from the list, or start typing its name. Only channels where the Kaiya bot is a member are listed, so add the bot to a channel first if you don't see it.
  * **Send to my DM:** sends the briefing to you in a direct message. A note warns that personal delivery may not work until you have started a chat with the Kaiya bot in Teams. Send the bot a message first.
* **Email Subject:** the subject line of the briefing email. It is filled in for you with **\[Kaiya]**, the Mission name and the frequency. Change it if you like. If it is empty, the hint text shows the usual pattern.
* **Message:** a short note filled in for you. Change it to give readers context, such as what to look at first. The same note goes with the briefing to every enabled channel.

The screenshot below shows the Slack and Microsoft Teams switches turned on, with **Send to my DM** selected for Teams.

<figure><img src="https://1702728550-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FC5PhRxmOjkQZJEO6F9Zw%2Fuploads%2FU5EYfqTBp1QUDzPpK7vW%2Fhd-teams-delivery.png?alt=media&amp;token=40ff0068-e5b6-4208-b497-856f5c2206f7" alt="Slack and Microsoft Teams turned on, with Send to my DM selected for Teams" width="563"><figcaption></figcaption></figure>

The bottom of the tab holds two buttons:

* **Reset:** resets the fields on the tab.
* **Create Schedule:** saves the schedule. The Mission then runs on its own from the start date. If the Mission has input steps, Kaiya then opens the schedule's conversation, runs the Mission up to each input step and asks you for the value there, or takes a hint on how to choose it, and reuses your answers on every run. See [Scheduling Missions](https://help.tellius.com/kaiya/kaiya-missions/scheduling-missions).

{% hint style="info" %}
To post briefings to Slack or Microsoft Teams, your workspace needs the Kaiya bot set up. See [Slack Integration](https://help.tellius.com/settings/administration/slack-integration) and [Teams Integration](https://help.tellius.com/settings/administration/teams-integration).
{% endhint %}

{% hint style="success" %}
You can also schedule by asking, for example *"Schedule it to run every Monday at 8 AM"* or *"Also send it to Slack."* Asking Kaiya to send the report to a colleague adds that person as a recipient. If Kaiya needs more detail, it asks a clarifying question. It then makes the change and confirms it in the conversation.
{% endhint %}


---

# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.

## Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation by asking a question.

Perform an HTTP GET request on the following URL with the `ask` and `goal` query parameters:

```
GET https://help.tellius.com/tellius-6.4/kaiya/kaiya-missions/how-to-create-a-kaiya-mission/the-mission-pane.md?ask=<question>&goal=<user_goal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is what the user is ultimately trying to achieve, the reason they need the answer. Sharing it helps GitBook give you a better, more relevant answer. A goal is most helpful when it describes the outcome the user wants rather than restating the question. For example, with `ask=how do I create an API token`, a goal like `automate deployments from our CI pipeline` lets GitBook tailor the answer to that use case.

The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.
