Perplexity launches Teammates AI agents for Max users

Perplexity has launched Teammates, a new AI agent feature for individual Max subscribers on the web. The feature is available as a research preview.

Teammates are designed to take ownership of ongoing work rather than handle one-off prompts. Users can assign a role such as monitoring bug reports, tracking metrics or researching a market, and the agent will plan tasks, carry them through and return the result for review.

The launch is limited in scope. Access requires an individual Perplexity Max subscription. Pro users are excluded, and Enterprise Pro and Enterprise Max are not included by default.

How it works

Each teammate has its own role and can work alongside the user and other teammates within the same project. Teammates can monitor connected tools, respond to changes and ask for user input when needed.

The product builds on Perplexity Computer, the company's existing system for tool use, files and code. This version adds support for ongoing tasks, larger projects and coordination across multiple agents.

When a user assigns a broader goal, a teammate can break the work into separate tasks, identify which can run in parallel and start additional agents to handle them. Each agent operates on its own computer and carries out its assigned task from start to finish before the results are collected and checked.

A teammate can continue monitoring a project after the work is completed and make updates based on what it finds. Output can include reports, pull requests or configuration changes for review.

Tool access

Teammates can connect to services including Slack, GitHub, Linear and Gmail through Perplexity's plugin system. Users can ask a teammate to connect a service from within chat, approve the requested permissions and then instruct it to monitor channels, repositories or workflows.

Teammates can also work with apps and websites through their own desktop computer and browser, including sites that do not have a dedicated connector. Users can take over and steer the work directly when required.

Within projects, teammates share context, files and instructions while keeping separate roles. They can also message each other when new information arrives, pointing to a more collaborative model of agent-based work within a single task environment.

Usage model

The service is tied to Perplexity's subscription and credit system. Max subscriptions include a monthly credit allocation that teammates can use, and more complex work may consume more credits, especially when several agents run in parallel, process large files or codebases, or run automations frequently.

Users can choose which model a teammate uses and can direct it to use different models for different tasks. New teammates use Claude Opus 5.5 by default, and any agents started by that teammate use the same model unless the user specifies otherwise.

Security limits

Teammates inherit the permissions of installed plugins and project settings. Saved credentials are encrypted and entered securely, while each teammate runs on an isolated computer with its own desktop, browser session and local cookies.

Project permissions continue to apply when teammates work inside a shared project. A teammate's chat is visible to its owner, while shared chats are read-only for others, although people can still work with a teammate in the Slack channels it monitors.

Perplexity's documentation sets out several intended use cases, including competitive research, issue investigation across GitHub, Linear and Datadog, and software optimisation work that tests several approaches in parallel and reports measured results.

The release underlines how AI companies are trying to move beyond chat-based assistance towards systems that can manage persistent assignments across multiple software tools while keeping a human review step before work is finalised.