Overview
Ekso includes a built-in AI assistant that can answer questions about your workspace, return structured data, and perform actions — all through natural language. The AI is available from the Ask tab in Insights. Ekso provides a default AI provider out of the box — you can optionally bring your own OpenAI or Azure OpenAI API key for direct API access.
The assistant draws on your workspace data and any documents marked with AI Context to provide relevant, grounded answers.
Chat interface
The AI chat interface sits at the bottom of the Ask tab. Type a question in the input field and click Ask or press Enter to submit it.
The toolbar next to the input provides:
| Button | Purpose |
|---|
| Ask | Submit the current question |
| + | Start a new conversation |
| History (clock icon) | Open conversation history |
| Info (i icon) | View sample questions |
Structured responses
The AI returns answers as formatted text, structured tables, or a combination of both. For example, asking “Show me most profitable items” returns a table with columns for Item, Item name, Container, Board, Billable hours, Cost, Revenue, and Profit — the same data you would see in the Item report, surfaced through a natural-language query.
Conversation history
Conversations are saved per user. Click the history button (clock icon) to open the History panel, which lists your past conversations with the opening question and date. Click a conversation to continue it, or click the delete button to remove it.
Each conversation maintains its full message history, so the AI can reference earlier questions and answers in the same thread.
What you can ask
The AI understands questions across your workspace. Open the info panel (i icon) to see sample questions, or type your own. You can specify date ranges, container names, board names, cycle names, and other filters in natural language.
Configuration and setup
Ask about how your workspace is configured:
- Which processes are accepted by the Fenrise container?
- Which fields are present in the Defect process as used by the Fenrise container?
- List workflow status values and their type
- Which process types exist?
- Show me the Defect process definition
- Check the workflow transition for the Defect process to see if it can be improved, given the status list for the system
Financial and profitability data
Query cost, revenue, and margin figures:
- Show me most profitable items
- Show me revenue, profit, cost numbers for November and December
- What is the mark-up breakdown for job roles?
- Which container brought in the most revenue?
- Show me revenue, profit, cost numbers for last month, for job roles that contain “Analyst” or “Developer”
Time and utilization
Explore how time is being spent:
- Get me hours logged by time code for this month
- Get time logged per week for the last two months
- Get hours logged, billable, non-billable per user
- Which job roles are doing the most work?
- Who was over-utilized last month?
- Who was not working to capacity last month?
- Which person logged the most hours where the job role was like “Developer”?
- Show me how many hours per day each user is scheduled to work?
Risk and delivery
Identify items and cycles at risk:
- Show me items where time logged > work estimate
- Which cycles are at risk of not getting all work completed on time based upon cycle finish date?
- Show me which boards have budget changes
- What cycles do we have due this month?
Item and billing summaries
Get item-level data:
- Show me top billable items
- Show me item billable summary
- Which resource that logged time has the highest base cost as per job role?
The AI can also perform actions in your workspace — not just answer questions:
- Create me a new board “Fred”
- Create new board called Dele with description “Hello World” and add three cycles
Actions performed by the AI follow the same access control rules as the rest of Ekso. You can only create or modify resources you have permission to manage.
Context documents
Documents in Docs can be marked with the AI Context toggle. When enabled, the document content is included as context when the AI answers questions — giving it access to your organization’s internal knowledge, runbooks, coding standards, or guidelines.
This is useful for grounding the AI in your team’s specific processes and terminology without needing to explain them in every conversation.
Enable AI Context on documents that contain stable reference material — architecture decisions, onboarding guides, and standard operating procedures work well. Avoid enabling it on frequently changing drafts where stale content could produce inaccurate answers.
Semantic search
AI search finds content by meaning rather than exact keyword match. Enter a phrase and Ekso returns results ranked by semantic similarity across items and documents. This is useful when you know what you are looking for but not the exact wording used.
Search results include the entity type (item or document), name, and a relevance score.
Providers
Ekso includes a built-in AI provider by default — AI features work without any additional configuration. If you prefer to use your own API key, you can configure one of two external providers under Settings > AI.
OpenAI
Connect directly to the OpenAI API.
| Setting | Description |
|---|
| Active | Enable or disable this provider |
| Model | The OpenAI model to use (for example, gpt-4o) |
| Key | Your OpenAI API key |
Azure OpenAI
Connect to an Azure-hosted OpenAI deployment. Use this if your organization requires data to stay within your Azure tenant or needs to comply with data residency policies.
| Setting | Description |
|---|
| Active | Enable or disable this provider |
| Model | The deployed model name |
| Endpoint | Your Azure OpenAI resource endpoint URL |
| Key | Your Azure OpenAI API key |
A provider is only active when all its required fields are filled in — toggling Active on without entering a key and model has no effect.
AI features are rate-limited separately from the REST API. If you see rate limit errors, wait a moment before retrying.
