Loading
PLATFORM
CAPABILITIES
ASANA AI
PLANS
Company Type
Teams
Industries
Use Cases
LEARN
SUPPORT
SERVICES
Help Center
search-icon

Topics

AI Teammates

In This Article

Note iconNote

Asana AI Teammates are currently in beta and are not yet generally available. During this period, features may change and you may experience bugs or performance issues. AI Teammates are free to use while in beta.

What are AI Teammates?

AI Teammates are collaborative AI agents embedded in Asana. They help marketing, IT, operations, and development teams deliver outcomes faster by working in the flow of your projects. Instead of acting as personal bots in a separate tool, they work where your team already collaborates, so everyone can see what is happening and contribute. 

AI Teammates are designed for teams, not just one person at a time. Multiple colleagues can assign work, ask questions, and review what the AI Teammate produces. Over time, they build shared memory that helps them understand your organization’s language, processes, and priorities.

They draw on the Asana Work Graph for context, using the projects, tasks, goals, and key resources you provide. They break down complex work into smaller steps, keep their activity visible in Asana, and follow the same access controls as any other user in your domain.

Here are some of the key ways AI Teammates work:

Context: AI Teammates understand your team’s work through the Asana Work Graph. They can use projects, tasks, goals, assignees, custom fields, and key resources to understand what you are asking them to do. You can refine their output by assigning work, giving feedback, and adding documents such as playbooks or style guides for them to consult.

Checkpoints: AI Teammates show their work inside Asana. They break complex work into subtasks, leave comments as they go, and update task fields so you can follow progress. Because their activity lives in the same tasks and projects as the rest of your work, your team can review drafts, request changes, or step in at any point.

Control: AI Teammates follow the same access model as users in your Asana domain. They can only see and use information that they have permission to access, and they reflect back what they are doing through visible activity in tasks, projects, and goals. You decide which teams, projects, and key resources each AI Teammate can use.

How to create an AI Teammate

You can create an AI Teammate using the + Create button in the top-left corner of Asana, the same way you would create a new task, project, or goal. Choose AI Teammate from the dropdown menu.

This will open the AI Teammate gallery. In the gallery, you can create an AI Teammate from scratch or start from a prebuilt option. To go directly to the AI Teammate gallery directly, click this direct link to the AI Teammate gallery.

Give your AI Teammate a clear name and description, then add it to a team. A descriptive name and short summary help colleagues understand what it is for when they see it in tasks and projects. Adding it to a team gives the AI Teammate access to that team’s work so it can provide more relevant help.

As AI Teammates work alongside you in Asana, colleagues will come across it. A unique name and a concise description will help others to recognize your AI Teammate and understand what it is used for. Adding the AI Teammate to a team gives it access to your team’s work so it can provide relevant help.

Visit your AI Teammates profile page to see an overview of its Tasks, Key resources, Behavior, and Memory. These are explained in more detail below.

On the AI Teammate profile page, you can see and configure several key areas.

Tasks

Tasks shows a running list of all the tasks assigned to the AI Teammate, both complete and incomplete. You can filter to Assigned by you to focus on work you have asked it to do.

Key resources

Key resources are documents and links you want the AI Teammate to consult every time it works. You might add a playbook, policy, glossary, or style guide so it can stay consistent with your team’s preferred language and processes.

Behavior

The behavior section guides how your AI Teammate collaborates, communicates, and makes decisions. You can describe what good output looks like, how formal or informal it should be, and how it should use tools such as comments, Google Docs, or Sheets.

Memory

AI Teammates store memories related to tasks they work on. These memories help them reuse context on future work, such as your preferences or decisions that should apply again, while still respecting task and project permissions.

Sharing

Use the sharing controls on the AI Teammate profile to decide who can use, edit, or manage it. Sharing controls which colleagues can assign tasks to the AI Teammate and which colleagues can change its behavior, key resources, and other settings

Using AI Teammates

AI Teammates collaborate like any other member of your team. You can assign them tasks, mention them in comments with questions or instructions, and watch as they respond and update work in context.

A simple way to get started is to pick a real task in a project you already use and assign it to an AI Teammate with a clear outcome in mind. For example, you might ask it to turn rough notes into a draft brief, review a set of tickets and summarize patterns, or structure a launch plan from an existing strategy document.

When you assign a task or @mention an AI Teammate, it will reply in the task. When helpful, it can also create or edit linked Google Docs or Sheets to hold longer drafts or analyses. You can review what it produced, respond with feedback or approvals, and let it iterate until the work is ready. Its subtasks and updates stay in sync so your team can follow along.

AI Teammates are a good fit for collaborative work that needs context and iteration. They can create content, brainstorm ideas, synthesize research, draft documents, analyze data, track progress, flag risks, and provide recommendations based on your team’s goals and plans.

AI Teammates quick-start guide

  1. In any task, assign or @-mention an AI Teammate with a clear ask and constraints.
  2. The AI Teammate will reply in the task and, when helpful, create or edit linked Google Docs or Sheets.
  3. Review the draft, reply with edits, or approve and finalize, and the AI Teammate will iterate and mark relevant subtasks complete.

Using prebuilt AI Teammates

You can find prebuilt AI Teammates in the AI Teammate gallery.

From there, you can browse options tailored to common types of work in marketing, product and engineering, and ops and PMO. Each prebuilt AI Teammate has a specific focus. You can start by assigning it a few simple tasks that match that focus, then refine its behavior and key resources as you go. Be sure to share the right context, such as relevant projects, portfolios, and documents with your AI Teammate so they can be at their most effective.

Below, we’ll give an overview of each prebuilt AI Teammate and some suggested tasks to assign to them.

Marketing AI Teammates

The marketing section of the gallery includes Content Strategist, Creative Partner, and Campaign Strategist. These AI Teammates are designed to help you shape strategy, create and refine assets, and guide campaigns from planning to reporting.

Content Strategist is designed to shape content strategy across channels, identifying high impact themes, mapping content to the buyer journey, and spotting gaps and opportunities.

You might create a task asking it to review your recent campaigns and “identify three high impact themes to focus on for next quarter,” or to “map our existing content library to the buyer journey for product X and list any gaps by stage.” Another useful task is to “analyze blog and email performance for the last six months and suggest opportunities where we are under-serving key audiences.”

Creative Partner helps drive ideas from draft to review by writing content outlines, generating variations, and reviewing deliverables.

As a starting point, you could assign a task like “draft an outline for a landing page about our new integration, based on this brief,” followed by another task to “generate ten headline and subheader variations for this campaign using our brand voice guide.” You can also ask it to “review this email sequence for clarity and tone and propose edits,” then decide which suggestions to keep.

Campaign Strategist guides campaigns from planning to reporting. It focuses on drafting campaign briefs, defining audiences and messages, and reporting on ROI and risks.

You might ask it to “draft a campaign brief for our upcoming webinar series using these goals and key resources,” or to “propose target audiences and core messages for a launch of feature Y, including an initial channel mix.” After a campaign ends, you can create a task like “summarize performance of the Q3 nurture campaign and highlight ROI, key learnings, and risks for next quarter.”

Product and Engineering AI Teammates

In Product & Engineering, you’ll see Spec Reviewer, Bug Investigator, and Sprint Accelerator. These AI Teammates are designed to help you check specs, analyze bug reports, and keep sprints on track.

Spec Reviewer checks specs for clarity and completeness by spotting missing details, flagging unclear requirements, and uncovering dependencies.

You can create a task such as “review this product requirements doc for our reporting improvements and list any missing details or unclear requirements,” or “read these linked specs and call out cross team dependencies we should track explicitly.” Another helpful task is to “scan the spec doc and comments and suggest questions we should answer before handing this to engineering.”

Bug Investigator analyzes bug reports and identifies blockers. It clarifies and clusters bug reports, assesses severity and impact, and surfaces critical bugs.

You might assign it a task like “review the last thirty days of bug tickets in this project and group them into themes, highlighting any high severity issues,” or “analyze these customer reported bugs and suggest which ones are likely related to the same root cause.” You can also ask it to “prepare a short summary of the most critical bugs affecting this release, including suggested next steps.”

Sprint Accelerator helps drive sprint goals, progress, and retrospectives. It reviews sprint goals, tracks progress and risks, and captures outcomes.

As a first step, you can create a task such as “summarize our current sprint goals, key stories, and any obvious risks based on this board,” or “prepare a mid sprint check in note using task status and comments in this project.” At the end of a sprint, you might ask it to “draft a sprint recap that captures what shipped, what slipped, and any follow up actions for the next sprint.”

Ops and PMO AI Teammates

For ops and PMO work, the gallery includes Launch Navigator, Strategic Planner, and Insights Analyst. These AI Teammates keep launches on track, turn inputs into actionable plans, and deliver updates for leadership decisions.

Launch Navigator keeps launches on track across teams by assessing launch readiness, monitoring dependencies, and detecting risks.

You can start with a task like “review this launch project and assess overall readiness, listing any risks or missing owners,” or “analyze dependencies across these tasks and suggest a simple checklist to track them.” Ahead of a key milestone, you might ask it to “prepare a pre launch health check that highlights blockers, timing concerns, and follow ups.”

Strategic Planner turns inputs into clear, actionable plans. It aligns plans to OKRs, models scenarios and forecasts, and highlights tradeoffs.

A common first task is “turn this leadership brief into a draft project plan aligned to our current OKRs, with phases and suggested owners,” or “compare these two proposed timelines and explain the tradeoffs and risks of each option.” You can also ask it to “outline a scenario plan for best, expected, and worst case outcomes for this program, including which metrics to watch.”

Insights Analyst delivers updates for leadership decisions by condensing project updates, summarizing portfolio health, and flagging risks and escalations.

You might create a task such as “pull a concise update for leadership from these three projects, focusing on progress, risks, and next steps,” or “summarize the health of this portfolio and call out any initiatives that appear off track.” Another useful task is to “draft talking points for our next steering committee based on the latest status updates and comments.”

Access control for AI Teammates

As you start assigning work to AI Teammates, it is important to understand what they can and cannot see. The following section explains how access works so you know when an AI Teammate will be able to help with a task or project.

When first created, AI Teammates can see tasks and projects public to the domain – just like new users. AI Teammates must be explicitly added as members or collaborators to private work graph objects (e.g., tasks, projects, teams) in order to see and search through them. 

When you assign a task or ask a question to an AI Teammate, the AI Teammate will only respond using information from tasks, projects, and other objects both you and the AI Teammate can access. AI Teammates can’t reveal info that you as the “triggering” user don’t already have permission to view.

For example, if you ask your AI Teammate to answer questions about a particular Asana project, they can only access and respond to you about that project if both of you have access to that project and/or its tasks. In other words, both you and your AI Teammate need to be authorized project members or task collaborators.

Details about access control

  • AI Teammates see public Asana data by default, like any new user in your domain (e.g., tasks in public projects; projects in public teams).
  • Requests involving or accessing information from third-party tools like Google Drive use the triggering user’s access tokens.
  • When creating or editing an AI Teammate, authorized members may upload Key Resources (e.g., playbooks, policies, instructions) for ongoing reference and access by the AI Teammate.

In practice, this means an AI Teammate cannot be used to look up information you do not already have access to. It can only work with the same tasks, projects, teams, and third party files that you and the AI Teammate can both see.

If you want an AI Teammate to help on a private project or team, make sure it has been added as a member or collaborator in the same way you would add a new colleague.

Memories

To stay helpful over time, AI Teammates can remember important details about the work they have done. This section explains how those memories work and how they respect the same access controls as your tasks and projects.

After an interaction involving a task, an AI Teammate may store a memory relating to the task for future reference. Think of memories as private comments or fields that only can be viewed by users who can both see the task and can trigger the AI Teammate.

Details about memories

  • Memories are stored on tasks as point-in-time recollections or summaries of the interaction that relates to a task.
  • A user who loses access to a task, will also lose access to its memories.
  • When an AI Teammate is removed from a project or task, the AI Teammate will forget the related memories.
  • Memories can be viewed by authorized members of the AI Teammate on the AI Teammate’s profile page.

Put simply, memories help an AI Teammate stay consistent with your preferences and past decisions, but they never widen access. Only people who can already see the underlying task, and who can work with the AI Teammate, can see its memories.

Over time, this lets teams build shared context with an AI Teammate so it can pick up work where it left off and avoid repeating the same questions, while still following your existing permissions.

AI Studio and AI Teammates

AI Studio and AI Teammates are designed to work together. AI Studio helps you build automated workflows for high volume, repeatable tasks. AI Teammates focus on collaborative work that needs context, judgment, and back and forth interaction.

One common pattern is to use AI Studio to collect and route requests into Asana as structured tasks, then involve an AI Teammate to handle the more complex work inside those tasks. For example, AI Studio might create and triage incoming campaign requests for a marketing team. A marketing AI Teammate can then draft briefs, summarize research, and prepare updates for those campaigns.

To learn more about AI Studio and how to build workflows, see the AI Studio articles in the Asana Help Center.

What AI Teammates can and can't do today

AI Teammates are still in beta and as such, we're adding more functionality all the time. We'll update this section as AI Teammates develop.

Invoking an AI Teammate

CapabilityCan an AI Teammate do this today?Notes
Trigger on task assignment✅
Trigger on @mention✅
Trigger on updates to objects that the Teammate is a collaborator on.✅If an AI Teammate is already a collaborator on a task or project, updates to that work (like task completion or key custom field changes) can trigger it to run again and decide what follow‑up actions to take.
Trigger via rule✅ (except for relative date rules and scheduled triggers)Because AI Teammates are collaborators, rules that assign tasks or move work to them can indirectly trigger executions. There is not yet a rich “AI Teammate action block” that lets you define arbitrary multi-step automations it must run.
Trigger upon form submission for Organization only forms✅
Trigger upon form submission for Anyone can access forms❌For access control safety, the AI Teammate can only run when it has user credentials from the form submitter.
Trigger via a recurring task❌AI Teammates can technically do the work on a recurring task, but it will not be able to mark the task complete.
Trigger via a relative due date rule (1 day before, 1 week after, etc.)❌

Searching the Work Graph

Tip iconTip

For best results, tell your AI Teammate where and what to search for in its behavior guidance or in your prompts.

CapabilityCan an AI Teammate do this today?Notes
Search public workspace data✅
Search tasks✅ - when it has access
Search subtasks✅ - when it has access
Search projects and project updates✅ - when it has access
Search milestones✅ - when it has access
Search portfolios✅ - when it has access
Search messages✅ - when it has access
Search comments✅ - when it has access
Search attachments✅ - when it has access
Search custom fields✅ - when it has access
Search status updates✅ - when it has access
Search goals and goal updates✅ - when it has access
Search reporting and resourcing❌
Search for objects it doesn't explicitly have access to within an interaction❌AI Teammates can only interact with information if both the AI Teammate and the running user have access. See Access control for AI Teammates.

Updating the Work Graph

CapabilityCan an AI Teammate do this today?Notes
Set name of task✅Can do for both existing tasks and tasks it created.
Mark a task complete✅Can do for both existing tasks and tasks it created.
Set description of task✅Can do for both existing tasks and tasks it created.
Add comments to a task✅Can do for both existing tasks and tasks it created.
Create tasks and subtasks✅AI Teammates can create tasks including titles, descriptions, assignees, and due dates.
Create subtasks✅ - only on tasks the AI Teammate has created
Edit tasks and subtasks✅Edit titles, descriptions, assignees, and due dates on tasks it’s working with.
Add comments to tasks and subtasks✅AI Teammates use comments to summarize work, propose next steps, and draft plans/status updates.
Delete tasks✅ - requires approval if it was not the task creator
Assign tasks or subtasks to a user✅ - requires approval if the user is not already a task collaborator or assigneeThe AI Teammate can assign a task to a user on its own if the user already has access to the task. If the user does not have access to the task, the AI Teammate will ask the running user for approval.
Update value of an existing custom field✅
Add task to a project✅ - requires approval if AI Teammate is not an editor in the project
Add or remove task collaborators✅ - requires approval
Create projects✅
Create or update sections✅
Create a custom field❌
Remove task from a project❌
Create dependencies❌
Bulk update tasks❌There is not yet a true bulk operation like “reschedule all 500 tasks by +2 weeks” or “bulk change a field across a project” today.
Create milestones or goals❌
Send messages or status updates❌
Create dashboards or charts❌
Add or remove users from Work Graph objects❌

External tools and files

CapabilityCan an AI Teammate do this today?Notes
Generate external files (documents, slides, and sheets)✅

The AI Teammate can create external files via Google Drive, Sharepoint, and OneDrive. 

The AI Teammate uses the running user’s authentication to access external tools. The running user must grant the AI Teammate access on the Profile page or when the request is triggered.

Search external files (images, PDFs, documents, slides, and sheets)✅ - if both the acting user and the AI Teammate have access

The AI Teammate can search external files via Google Drive, Sharepoint, and OneDrive. 

The AI Teammate uses the running user’s authentication to access external tools. The running user must grant the AI Teammate access on the Profile page or when the request is triggered.

Generate images or PDFs❌
Attach a file to a task❌Teammates can link, but not yet attach.

Access, safety, and usage

CapabilityCan an AI Teammate do this today?Notes
Bypass permissions❌

 

Related articles

Loading
Asana AI Teammates