> For the complete documentation index, see [llms.txt](https://docs.nimbalyst.com/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.nimbalyst.com/task-management/ai-integration.md).

# AI Integration

By default, AI agents in Nimbalyst can use the tracker system. They can create, update, and query items during a session, and you can link the ones that matter back to the session for traceability. You can turn this access off per project.

### Creating Items

During a coding or planning session, the AI can create tracker items on the fly when it identifies work to track:

> "I found a potential memory leak in the WebSocket handler. I've created a bug item to track it."

You can also ask the agent directly:

> "Create a task for updating the API documentation" "Log an idea: real-time collaboration for the editor" "Track a decision: we chose Tailwind over styled-components"

The agent selects the appropriate type, sets fields based on your description, and creates the item.

### Updating Items

Ask the agent to modify existing items:

> "Mark the login bug as done" "Set the database migration plan to in-progress and assign it to me" "Add a comment to bug #123 saying the fix is in PR #456" "Change the priority of the caching task to critical"

The agent can update any field: status, priority, owner, description, tags, and progress.

The agent can also **reclassify an item to a different type** — for example, turning a task into a bug — without losing its comments, attachments, or session links. Just ask:

> "Convert task #42 to a bug" "This idea is actually a feature, reclassify it"

### Querying Items

Ask about your project's current state:

> "What bugs are currently open?" "Show me all high-priority tasks assigned to me" "List decisions made this month" "How many items are in the In Review column?" "Find all items tagged with 'backend'"

The agent can filter by type, status, priority, owner, tags, and perform full-text search across titles and descriptions.

### Session Linking

When an agent creates or updates a tracker item with an explicit link (via `linkSession: true` on `tracker_create`, or by calling `tracker_link_session`), the item is connected to that session. This means:

* From any item, you can see which sessions touched it
* From any session, you can see which items were created or modified
* You can trace the full history of an item back to the conversations that shaped it

Neither creating nor updating an item auto-links the current session. The agent has to opt in (for example with `linkSession: true`), which keeps sessions from accumulating unrelated items when the agent is just logging work in passing.

### AI Agent Access

You control whether AI agents can use your trackers, per project. In **Settings > Project > Trackers**, the **AI Agent Access** toggle turns the tracker tools on or off for that project. It is on by default. Turn it off and the tracker tools are removed from the agent entirely, so it can neither read nor change your items. The change takes effect for new agent sessions.

### Transcript Widgets

Tracker operations performed by the agent appear as structured widgets in the chat transcript, not just plain text. Each widget shows the item type, title, status, and other key fields. Click the widget to navigate directly to the item in the tracker.

This makes it easy to review what the agent created or changed without scrolling through conversation text.


---

# 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 dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter, and the optional `goal` query parameter:

```
GET https://docs.nimbalyst.com/task-management/ai-integration.md?ask=<question>&goal=<endgoal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is optional and describes the broader end goal you are ultimately trying to accomplish on behalf of the user. GitBook uses it to tailor the answer towards what is most useful for that goal.

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.