ProductPerplexityConfirmed
Story 9 of 16 in the Thursday 19 November editionPerplexity adds self-improving skills to Computer Brain
Rolled out in Perplexity Computer on 19 November 2026 after an eight-week production trial that lifted personal skill adoption to 72.9% and cut tool errors 11%.

Perplexity updated Brain, the personalised memory and workspace-wiki agent inside Perplexity Computer, on 19 November 2026 to automatically author and maintain self-improving agent skills from each user’s completed work, extending the platform’s recent Multiagent Computer and Teammates rollout. An internal improvement agent inspects full session trajectories covering tool calls, reasoning steps and final outcomes, proposing a new personal skill or revising an existing one only after recurring patterns establish clear evidence.
Paired failure-replay review and eight-week randomised production trial
While personal skills adapt to individual workflows, shared built-in skills distributed to every Perplexity Computer user must pass paired replay against historical failure scenarios plus human engineer review before any update ships to production.
Chief executive Aravind Srinivas published metrics from an eight-week randomised controlled trial on live Perplexity Computer traffic: 72.9% of users in the treatment cohort received at least one auto-generated personal skill versus 6.5% who added a skill manually, reducing tool-call failures by 11% and dissatisfied threads by 7.6% with zero measurable increase in cost or latency.
Key details
- Agent surface and scope
- Brain inside Perplexity Computer (personal skills + platform-wide built-in skills)
- Personal skill adoption (8-week RCT)
- 72.9% of users gained >= 1 personal skill vs 6.5% manual creation
- Production reliability impact
- 11% fewer tool-call failures and 7.6% fewer dissatisfied threads; zero latency/cost change
Release details & context
Evidence gating for personal skills versus engineer-reviewed built-in skills
Perplexity separates individual customisation from global skill updates using two distinct promotion gates. For single-user workflows, Brain builds atop the knowledge wiki it already maintains about a user’s projects and withholds skill writes until repeated trajectory traces confirm a recurring pattern.
For global built-in skills shared across all Perplexity Computer accounts, candidate edits cannot go live autonomously: every change is benchmarked via paired replay on recorded failure cases and signed off by a Perplexity engineer.
Perplexity release timeline (14 launches tracked)
- Perplexity launches Frontier Research lab backed by 10,000+ GPUs
Perplexity forms Perplexity Frontier Research with 10,000+ GPUs to build autonomous research systems and publish open-source models and data.
- Perplexity launches Orchestrator API to route across 45+ AI models
Perplexity launches its Orchestrator API to route requests across more than 45 open-weight and proprietary models under one API key and bill.
- Perplexity rolls out Multiagent mode in Computer and Teammates for Max
Perplexity rolls out a Multiagent effort level on web for Max subscribers in Computer and Teammates to run parallel multi-model research streams.
- Perplexity launches Context API for agent URL extraction at $1 per 1k
Perplexity ships its Context API at $1 per 1,000 successful URLs, hitting 92.3% usable extraction and 87.9% fact recall across 715 sites.

