> For the complete documentation index, see [llms.txt](https://docs.usepylon.com/pylon-docs/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.usepylon.com/pylon-docs/account-intelligence/tasks-and-projects/smart-projects.md).
# Smart Projects
Smart Projects use AI to keep a project's tasks current as the customer account changes. Pylon evaluates relevant account activity and can create or update tasks based on the AI instructions you provide.
Use Smart Projects for workflows that evolve over time, such as customer onboarding, technical implementations, escalations, and renewal-risk recovery.
{% hint style="info" %}
A project becomes a Smart Project when it has AI instructions. Without project-level AI instructions, it behaves like a standard project.
{% endhint %}
### How Smart Projects work
1. Add AI instructions to a project or Project Template.
2. When a project is created from a template, it inherits the template's AI instructions.
3. Pylon evaluates the project when it is created and when relevant activity is recorded for its account.
4. Task changes and their supporting reasoning appear in the project's activity log.
New activity does not always result in a change. Pylon leaves the project unchanged when there is not enough evidence for a specific update.
### What AI can do
Smart Projects can automatically:
* Create a task when new work is not covered by an existing task
* Update a task's title, description, AI instructions, assignee, due date, or milestone
* Change a task's status when account activity supports the change
Smart Projects do not update project or milestone details or mark the project itself complete.
For an individual task, AI can also prepare these actions for a teammate to review:
* A Slack notification to an available account channel
* A draft for a new customer issue
* An update to a supported task or account field
{% hint style="info" %}
Prepared actions are never sent or applied without human approval. This is enforced by Pylon and does not depend on wording such as "keep customer-facing updates human-reviewed" in your instructions.
{% endhint %}
### Configure a Smart Project
AI instructions tell Pylon how to maintain the project's tasks. Specify:
* When a task should be created or updated
* Which owner, milestone, status, or due date to use
* What evidence permits a status change
* What task or review action should be created when progress is blocked
For example:
> When the customer confirms a launch date, update the due dates of the Launch milestone tasks. If required setup remains incomplete seven days before launch, create a task for the account owner to review the blocker.
Use project-level instructions for behavior that applies across the project. Add task-level instructions only when one task requires different evidence or a specific review action.
Instructions do not add new data sources, enable additional actions, or change Pylon's approval requirements.
### Create a Smart Project Template
1. Go to [Project Templates](https://app.usepylon.com/settings/project-templates) in Settings.
2. Create a template or open an existing one.
3. Expand **AI instructions** and describe how Pylon should maintain projects created from the template.
4. Add task-level AI instructions where a task needs more specific guidance, then save the template.
Projects created from the template inherit its AI instructions. You can edit the instructions on an individual project without changing the original template.
### Review AI activity
Open a Smart Project's activity log to review task changes and the evidence used to support them. Open an individual task to review any actions waiting for approval.
---
# 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.usepylon.com/pylon-docs/account-intelligence/tasks-and-projects/smart-projects.md?ask=&goal=
```
`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.
