What it means
The claw metaphor describes how autonomous layers extend from chat into systems of record. Users feel it as background work, spawned sub-agents, and actions outside the main conversation thread.
Why designers should care
Claw-style autonomy needs visible scope, spawned-session status, completion reports, and the same approval patterns as tool use. Silent background reach erodes trust after one wrong edit.
Example
A coding orchestrator spawns a child session to run tests; the UI shows “Spawned session: fixing lint errors” with a link to logs and a single Approve merge button when done.
Common mistakes
- Spawned agents that hide actions until after irreversible changes.
- No difference in UI between chat replies and environment-changing claw work.
- Parent and child sessions with conflicting permissions and no audit trail.
Explore more
Frequently asked questions
What is Claw?
In agent tooling, a claw is the reach of an AI orchestrator into your environment: spawning sessions, running commands, editing files, or driving browsers on your behalf.
Why should designers care about Claw?
Claw-style autonomy needs visible scope, spawned-session status, completion reports, and the same approval patterns as tool use. Silent background reach erodes trust after one wrong edit.
What are common mistakes with Claw?
Spawned agents that hide actions until after irreversible changes; No difference in UI between chat replies and environment-changing claw work; Parent and child sessions with conflicting permissions and no audit trail.