Glossary
The terms used around Helix.
Helix glossary — short, factual definitions for the terms that show up around agent identity, MCP, PASETO, predictive UI, voice-first AI, and the approval loop. Linked back to the product surfaces where each idea is materialised.
Agentic identity
The bundle of permissions, addresses, calendars, contacts, and approval rules an AI agent needs to act on a user’s behalf.
MCP (Model Context Protocol)
An open standard that lets LLMs call typed, schema-described tools over a single endpoint.
PASETO
A modern alternative to JWT for signed, short-lived authentication tokens with safer defaults.
Predictive UI
A UI pattern that surfaces the next likely action before the user asks for it.
Voice-first AI
A product pattern that treats voice as the primary input with typed input as the fallback.
Approval loop
The human-in-the-loop pattern where every agent send queues for explicit human approval before going out.
Human in the loop
A design pattern in which an AI agent pauses for explicit human approval on consequential actions — sends, calendar writes, destructive operations — instead of acting fully autonomously.
Lethal trifecta
Access to private data, exposure to untrusted content, and the ability to communicate externally — combined in one agent, this is the prompt-injection attack surface for AI tools.
In product