Research Radar · Experimental implementation
Agent Shell
A collaborative terminal interface for concurrent AI-agent execution.
What it investigates
Agent Shell explores a shared terminal session where multiple agents can coordinate visible work and retain a recoverable history.
Implementation record
The README describes an event-sourced SessionManager, snapshots for recovery, file-edit records with hashes and ranges, a capability registry, task matching, conflict detection, and approval-oriented interaction. These pieces separate session history from the current reconstructed state.
Public record
The linked research item is the public source recorded for this Research Radar build. It provides context for the implementation; this page does not claim to reproduce its results. The repository and its documentation are the current public materials for inspecting the build itself.
Scope and limitations
This describes an experimental interface and its modules. It does not claim safe concurrent operation in every shell environment, complete conflict prevention, or production readiness.