graff — a fast agentic coding harness in Zig: multi-provider, MCP, workflows, DGM evolution loop, TS/Python SDKs
0026. Foreground bash auto-backgrounds after 120s, it is not killed
Status: accepted 2026-08-25
Context
#93 gave subagent bash a
120s kill because those children have no TTY. The root was left unbounded:
"a human is watching and may want a long build." #607 added a dim
· bash still running · pulse so a long command did not look hung.
#620 measured the leftover: a durability drill sat in that pulse for 83 minutes. The model never woke. Esc killed the process group; the model retried the same foreground hang.
xai-org/grok-build BashTool waits
120s (timeout default), then promotes a still-running command to a
background task (auto_background_on_timeout) and returns a job id plus
partial output. A real compile keeps running. The model gets the turn back.
Decision
- Root foreground
bashwaits up to 120s. An explicittimeoutmay shorten that wait; a larger value cannot extend it (#850). Process lifetime after promotion is the job registry, not the overloadedtimeoutfield. - If the child is still running, it is moved onto the existing job registry.
The model receives a job id.
bash_output/bash_kill//jobswork as they do forrun_in_background: true. The process is not killed. - Subagent
bashstill kills at 120s (#93). Children have no job UI. - Esc during the foreground wait still kills the process group. The cancelled
result tells the model to restart with
run_in_background: true. - ADR 0010 is unchanged:
bash_output(wait_ms>0)waits for exit, not a poll.
Consequences
A 20-minute compile or a forgotten server no longer bricks the turn. The
heartbeat (#607) still fires during the 120s wait. Root bash output that
finishes inside the wait is interleaved stdout+stderr (the job buffer),
not the old split [stderr] block — same shape as bash_output.