Perplexity's agent now writes and updates its own skills from experience
Perplexity has extended Brain, the personalised self-improvement system behind its Computer agent, so it can automatically create and update agent skills, reusable instructions that guide the agent through recurring tasks.
Brain uses a user's memory and session history to write and refine personal skills for that user's workflows. A similar process, based on patterns across all sessions, updates the built-in skills available to every Perplexity user. Perplexity says personalisation changes the agent's context and instructions, not the model's weights.
Each update runs through four stages: Brain finds past sessions where a skill fell short, an improvement agent proposes changes, checks confirm they are likely to help, and the verified version goes live. For built-in skills the bar is higher: changes must pass paired replay tests against the current version and only take effect once an engineer approves them.
Perplexity says Brain wrote more than 80,000 personal skills for over 20,000 users between 1 August and 14 November, and applied more than 100,000 automatic updates to over 30,000 existing skills since September. In an eight-week randomised test on production traffic, 72.9% of users with the feature enabled got at least one new personal skill, against 6.5% of a control group who added one manually. Software engineering was the largest category, at 11% of skills, according to Perplexity.
We hope you enjoyed this article
Consider subscribing to one of our newsletters like Daily AI Brief.
Also, consider following us on social media:
Daily AI Brief
Daily report covering major AI developments and industry news, with both top stories and complete market updates
Industry analysis
2025 Global Business Services Agenda: Gen AI Takes Center Stage
This industry analysis by The Hackett Group explores the transformative impact of generative artificial intelligence (Gen AI) on global business services (GBS) in 2025. The study highlights the shift from exploration to acceleration of Gen AI initiatives, with 89% of executives advancing these projects to improve customer satisfaction, innovate products, and reduce costs. The report also discusses the challenges and strategies for successful Gen AI adoption, emphasizing the need for a technology-enabled operating model and the importance of reskilling the workforce.
Read moreYou may also like
Perplexity launches Teammates, agents that own ongoing work inside Computer
Perplexity Launches Persistent AI Teammates for Max Subscribers
Perplexity launches Frontier Research to build an automated scientist
Perplexity tops $1 billion annualized revenue, CEO says, and may file for IPO as soon as early 2027
Perplexity Brings Fully Local Computer Workflows to Mac
Daily AI Brief: the AI news that matters, in your inbox.