Bug

When triggering a cron job manually via the cron run API/tool, the call blocks until the gateway times out (60s). The job still runs in the background, but the caller gets a timeout error even though nothing actually failed.

Expected Behavior

cron run should enqueue the job and return immediately with an acknowledgment (e.g. {ok: true, ran: true}). The caller doesn't need to wait for the job to finish or even start executing. It's a trigger, not an RPC.

Current Behavior

The API call hangs for 60 seconds, then returns a gateway timeout error. The job does eventually run, so this is purely a UX/latency issue on the trigger side.

Impact

• Wastes 60 seconds of tool call time for the agent
• Returns a misleading error (timeout suggests failure, but the job ran fine)
• Blocks the main agent session unnecessarily