DevOps / SRE / Platform · 21.08.2026, 13:16 UTC
Grok, Claude, and Hermes agents get job titles — and persistent permissions
| Schweregrad | info |
|---|---|
| Kategorie | DevOps / SRE / Platform |
| Quelle | The New Stack ↗ |
| Veröffentlicht | 21.08.2026 UTC |
Sicherheitsmeldung mit Schweregrad noch nicht bewertet. Technische Details im Tab „Originaltext“; empfohlene Schritte in der Checkliste.
Your next AI coworker may look like a chatbot, but underneath their name, face, and job title will be something more consequential: A persistent identity that uses memory, keeps permissions, and has responsibilities across all the conversations you have with it.
We can already see that model taking shape in Hermes. This month, Nous Research folded Bot Mode into the Hermes desktop app. The standalone plugin repository is archived; the roster now ships bundled and enabled by default; and v0.20.4 added a tabbed Sessions and Bots sidebar. Each profile shows up as a named bot with its own chat, avatar, personality, schedule, and pinned model.
Hermes tabbing Sessions beside Bots reflects a broader shift. Sessions did not disappear from any of these products; they sit under a durable agent identity that carries memory and permissions across them.
Hermes and Grok build rosters, Claude joins one
xAI announced Grok Bot on August 11, featuring a sidebar of named Bots, each with a title, a description, and up to 50 routines per account. Hermes now does the same thing on the desktop under an MIT license. Claude Tag from Anthropic, announced on June 23, takes the other route. One organization-provisioned Claude joins the human roster in Slack, where anyone in a channel can tag it, and each thread still spins up its own working session.
Three different products and deployment models took the same direction. The reusable object shifted from being the conversation to serving as the worker identity above it.
A bot is a profile, not a new runtime
Hermes is the easiest to read because …