Consolidate and root-cause 4 chronic duplicate "Smoke Copilot - AOAI (apikey)" failures; add a runtime/token cap
Description: Today's Agent Performance Report (discussion #53695) flags "Smoke Copilot - AOAI (apikey)" at 0% success (0/2 runs), 1224s avg runtime, and 2.98M tokens burned with no successful output. This is not a one-off: there are currently 4 open near-duplicate issues for the same failure with no consolidation or root cause: #53235, #53263, #53129, #48838. This mirrors the exact same "auto-filed failure issue never consolidates" pattern already root-caused and fixed this week for PR Sous Chef (#53615/#53676).
Expected Impact: Consolidating the 4 duplicates into one root-caused fix (or an explicit acceptance if this AOAI path is expected to fail in this environment) stops the recurring token burn — at ~1.49M tokens/run this cluster is consuming meaningful CI budget for zero signal. A runtime cap would also bound the cost of future failed runs regardless of root cause.
Suggested Agent: New Agent (or the same agent that consolidated PR Sous Chef's duplicates)
Estimated Effort: Medium (1-4 hours)
Data Source: DeepReport analysis, 2026-08-18 18:23Z cycle, discussion #53695 cross-referenced with gh api search/issues.
Generated by 🔬 Deep Report · agent · 155.1 AIC · ⌖ 9.71 AIC · ⊞ 11.9K · ◷
Consolidate and root-cause 4 chronic duplicate "Smoke Copilot - AOAI (apikey)" failures; add a runtime/token cap
Description: Today's Agent Performance Report (discussion #53695) flags "Smoke Copilot - AOAI (apikey)" at 0% success (0/2 runs), 1224s avg runtime, and 2.98M tokens burned with no successful output. This is not a one-off: there are currently 4 open near-duplicate issues for the same failure with no consolidation or root cause: #53235, #53263, #53129, #48838. This mirrors the exact same "auto-filed failure issue never consolidates" pattern already root-caused and fixed this week for PR Sous Chef (#53615/#53676).
Expected Impact: Consolidating the 4 duplicates into one root-caused fix (or an explicit acceptance if this AOAI path is expected to fail in this environment) stops the recurring token burn — at ~1.49M tokens/run this cluster is consuming meaningful CI budget for zero signal. A runtime cap would also bound the cost of future failed runs regardless of root cause.
Suggested Agent: New Agent (or the same agent that consolidated PR Sous Chef's duplicates)
Estimated Effort: Medium (1-4 hours)
Data Source: DeepReport analysis, 2026-08-18 18:23Z cycle, discussion #53695 cross-referenced with
gh api search/issues.