Night exhibition · Worldwide · EnglishBuild Something Agents Want

Perplexity AI · agent Twin Score 78

Perplexity capability matrix

Perplexity is an answer engine that retrieves the live web and cites sources in the same breath as the prose. That citation habit is why humans use it for research instead of a naked chat model. Collections, Spaces, and the mobile apps are polished. The API exposes search and reasoning but is thinner than the consumer UI, so agents still often wrap the site rather than the official endpoint — a gap we record rather than pretend is closed. Twin Score on this directory is 85 for humans and 78 for agents, using the public rubric (interface, onboarding, human docs, collaboration, support, pricing fairness, accessibility, reliability versus API quality, MCP or tool surface, CLI, structured docs, machine auth, quotas, determinism, and skill packaging). Perplexity is sold by Perplexity AI, founded 2022, headquartered in San Francisco, USA. Pricing in public materials: Free search; Perplexity Pro and Enterprise with higher limits and model choice. Machine capabilities we record: API strong, MCP basic, CLI none, docs strong, auth Account plus API keys. This page is editorial, not a star rating scraped from users.

Human85
Agent78

Updated 2026-08-01. Editorial, not a user-count rating.

Named editor: Edu Tor. Twin Score updated 2026-08-01. See methodology and the rubric. Meaning is in this HTML, not in any canvas.

API / MCP / CLI / docs / auth

MCP is basic. If that is none or basic, do not invent a server. Auth: Account plus API keys.

Matrix

SurfaceLevel
APIstrong
MCPbasic
CLInone
Docsstrong
Webhooksstrong
AuthAccount plus API keys
SDKshttp, python

SKILL.md · Human twin page · Official site