[deep-report] Add driver-level timeout/log signal for stuck Execute CLI steps #53179

Description

This week's Agentic Workflow Audit (discussion #52970, 2026-08-15) found a hang-type failure that was hard to diagnose: Daily Ambient Context Optimizer got stuck in an in_progress Execute CLI step with all downstream steps left pending — a silent hang, not a clean crash. The auditing agent could only distinguish this from a healthy run by manually cross-referencing job_details for stuck-pending downstream steps; there was no direct timeout/log signal marking it as a hang.

The same audit's Phase 3 recommendation: "Surface a clearer timeout message for stuck Execute CLI steps... a driver-level timeout with a clear log line would make this diagnosable without cross-referencing job_details."

Expected Impact

Makes hung Execute CLI steps (across engines) self-diagnosable from the step log alone — no more cross-referencing job/step metadata to distinguish "hung" from "still working" or "cleanly crashed." Should reduce time-to-diagnosis for this failure class and make automated audit/log-analysis workflows (like Agentic Workflow Audit itself) more reliable.

Suggested Agent

New Agent (harness/driver-level observability improvement)

Estimated Effort

Medium (1-4 hours)

Data Source

DeepReport analysis, 2026-08-16 cycle, sourced from discussion #52970 (Agentic Workflow Audit, 2026-08-15 run).

Generated by 🔬 Deep Report · agent · 257.9 AIC · ⌖ 11 AIC · ⊞ 11.8K · ◷

• expires on Aug 18, 2026, 10:35 AM UTC-08:00