Perplexity Launches Automations: A New Form of AI-Powered Workflow Execution
Perplexity adds Automations to its Computer product, moving AI from passive Q&A to proactive workflow execution.
Perplexity has launched Automations for its Computer product, marking a strategic shift from search and Q&A toward a productivity agent platform. The feature supports both event-based triggers and scheduled execution, enabling continuous task completion without manual intervention. Crucially, Automations integrates with user memory, a skills system, and third-party apps including Slack, Gmail, Outlook, and Linear — allowing AI to operate seamlessly across everyday work and development workflows. The move aligns with the broader industry trend toward agentic AI, though its real value depends on trigger accuracy and cross-app execution reliability.
Perplexity Computer Gets Automation Capabilities
Perplexity recently announced the introduction of Automations into its Computer product, shifting the AI assistant from a passive, on-demand responder to a proactive executor. This update targets continuous, repetitive work scenarios, enabling AI to complete tasks without requiring users to issue instructions each time.
According to the official announcement, Automations is designed for "ongoing work." Rather than being limited to back-and-forth interactions, it can take action based on event-based triggers or run on a schedule. This means users can define rules and let the system operate automatically whenever specific conditions are met.
Two Trigger Mechanisms: Event-Driven and Scheduled
The core capability of Automations lies in its two triggering modes.
Event-Based Triggers
Automated workflows are activated when a specific event occurs — for example, receiving a particular email or a project status update — prompting the system to execute a corresponding action. This mechanism suits workflows that require immediate response, turning AI into an always-on listener.
On a Schedule
The other mode runs according to a preset timetable — such as summarizing information at a fixed time each day or generating reports on a recurring basis. This type of automation is well-suited for regular, cyclical tasks, eliminating the need for users to repeat operations manually.
The combination of both mechanisms allows Automations to handle both unexpected events and routine recurring work, covering the most common automation use cases.
Deep Integration with Memory, Skills, and Third-Party Apps
What makes Automations particularly noteworthy is that it doesn't operate in isolation — it works in conjunction with several of Perplexity's existing capabilities.
The company explicitly states that Automations can leverage the user's memory and skills, and work in concert with connected third-party applications. Currently supported apps include Slack, Gmail, Outlook, and Linear.
This integration lineup signals a clear positioning intent: Slack and Gmail/Outlook cover team communication and email management, while Linear points to software development and project management. In other words, Perplexity wants Automations to serve as an automation hub spanning everyday work, team collaboration, and development workflows — not merely a search-enhancement tool.
With memory, automated workflows can retain context and accumulate user preferences. With the skills system, they can invoke predefined processing logic to handle more complex tasks. Together, these three elements enable a single automated workflow to execute coherently across applications and scenarios.
About Linear: Linear is a project management tool designed for software development teams, known for its clean interface and Issue-centric workflow. It is widely used among tech startups and engineering teams and is often regarded as a lightweight alternative to Jira. Perplexity's choice to integrate Linear rather than Jira or Asana reflects its primary target audience — teams with a strong technical background who prefer modern toolchains. Notably absent from the current integration list are knowledge management tools like Notion and Confluence, as well as more mainstream development collaboration platforms like GitHub and Jira. This somewhat limits Automations' applicability in broader enterprise environments. The scope of future integrations will directly determine how deeply this feature can reach across different user bases.
Industry Signal: AI Shifts from Q&A to Agentic Execution
Perplexity's move reflects a clear trend in AI product evolution — the transition from question-answering assistants to agentic executors. Over the past year, major AI platforms have increasingly emphasized that models should not only "know how to answer" but also "know how to act" — autonomously calling tools, connecting to external systems, and completing end-to-end tasks.
Automations is a concrete realization of this direction: by combining triggers, memory, and app integrations, it embeds AI capabilities directly into users' real workflows. For Perplexity, long known for search and Q&A, this represents a meaningful push into productivity tooling.
That said, the official information disclosed so far is fairly limited. Details around configuration, pricing, plans for expanding supported applications, and the stability of real-world execution remain to be seen. The practical value of automation ultimately depends on the accuracy of trigger detection and the reliability of cross-app execution — both of which will need to be validated through actual use.
About Agentic AI: "Agentic AI" refers to AI systems capable of autonomously decomposing goals, planning steps, calling external tools, and iterating through execution until a task is complete — distinct from traditional single-turn or multi-turn conversational models. In agentic mode, AI no longer waits for step-by-step user instructions; instead, it breaks down a high-level objective into a series of subtasks and dynamically adjusts its strategy based on feedback during execution. OpenAI's Operator, Anthropic's Claude with tool use, and Google's Project Astra are representative explorations in this direction. The technical foundations of this paradigm include Function Calling, Tool Use, and persistent memory — together giving models a closed-loop "perceive–decide–act" capability. For users, the core value of agentic AI lies in freeing people from repetitive "execution-layer" work so they can focus on goal-setting and outcome review. However, this also introduces new risks: autonomous execution means errors can be amplified without human oversight, making execution reliability and permission boundary design critically important.
Summary
Perplexity Computer's Automations feature adds a proactive execution dimension to the AI assistant. Through event-based triggers and scheduled execution, combined with memory, skills, and integrations with Slack, Gmail, Outlook, and Linear, it positions AI at the center of continuous workflows. This is yet another example of AI products evolving from passive responders to proactive agents — though its real-world effectiveness still awaits validation from broader user feedback.
Related articles
Chain of Thought Isn't Just for AI: How Humans Can Use CoT to Improve Thinking Quality
Chain of Thought isn't just a prompt trick for large models. Starting from a Hacker News discussion, this article analyzes why CoT works for human thinking too, breaks down actionable ways to transfer it, and highlights the key differences between human CoT and machine CoT.
Cloudflare Teams Up with OpenAI: Closi…
Cloudflare Teams Up with OpenAI: Closing the Loop from Vulnerability Discovery to Defense with AI
Cloudflare uses OpenAI Daybreak to investigate security vulnerabilities and generate reviewable remediation plans, while developers use Codex to boost coding efficiency and quality.
Orca Multi-Agent Orchestration in Practice: A Guide to Parallel Collaboration with Claude and Codex
Orca is Stable AI's free open-source agent workbench supporting 34 agents like Claude Code and Codex in parallel. Learn its Git worktree isolation, multi-model orchestration, and mobile management.