jackin eject
Synopsis
Section titled “Synopsis”jackin eject <SELECTOR> [OPTIONS]Stop a running agent container.
Arguments
Section titled “Arguments”| Argument | Required | Description |
|---|---|---|
SELECTOR | Yes | Role selector, instance ID, or container name |
Options
Section titled “Options”| Option | Description |
|---|---|
--all | Stop every matching instance. Without --all, the command errors if more than one instance matches the target |
--purge | Also delete persisted state after stopping |
--debug | Print every external command jackin issues and their captured output. Also enabled by JACKIN_DEBUG=1. |
Examples
Section titled “Examples”# Stop a specific agentjackin eject agent-smith
# Stop all instances of agent-smithjackin eject agent-smith --all
# Stop and delete all persisted statejackin eject agent-smith --purge
# Stop a specific indexed instancejackin eject k7p9m2xq
# Stop a specific container by namejackin eject jk-k7p9m2xq-agentsmithWhat gets cleaned up
Section titled “What gets cleaned up”When you eject an agent:
- The agent container is stopped and removed
- The DinD sidecar container is stopped and removed
- The TLS certs volume (
{container}-dind-certs) is removed - The per-agent Docker network is removed
When you add --purge:
- All of the above, plus the agent’s persisted state (conversation history, forwarded credentials,
ghstate, plugins) is deleted
Use --purge when Docker resources still exist and you want to delete local recovery state in the same operation. Plain jackin purge is local-state-only and refuses while the matching role container or DinD sidecar still exists.