Description
Three daily audits — Safe Output Health, API Consumption, and Agentic Workflow Audit — repeatedly compute their metrics on partial datasets because the logs MCP tool hits a 120s fetch cap (e.g. the audit ran on 100 of 204 downloaded runs; safe-output health on 82/131 conclusions). Fix #45510 was closed but the symptom persists, so daily health and cost numbers are computed on incomplete data.
Expected Impact
Pagination / streaming or a raised fetch cap for the logs tool would let these audits see the full daily run set, making fleet health and API-cost metrics trustworthy.
Suggested Agent
MCP / gh-aw infra agent — add pagination or increase the 120s fetch window in the logs bridge.
Estimated Effort
Medium (1-4 hours)
Data Source
DeepReport Intelligence Briefing — 2026-07-20. See Workflow Audit (#46683), Safe Output Health (#46714), API Consumption (#46808), prior #45510.
Generated by 🔬 Deep Report · 165.1 AIC · ⌖ 17 AIC · ⊞ 10K · ◷
Description
Three daily audits — Safe Output Health, API Consumption, and Agentic Workflow Audit — repeatedly compute their metrics on partial datasets because the
logsMCP tool hits a 120s fetch cap (e.g. the audit ran on 100 of 204 downloaded runs; safe-output health on 82/131 conclusions). Fix #45510 was closed but the symptom persists, so daily health and cost numbers are computed on incomplete data.Expected Impact
Pagination / streaming or a raised fetch cap for the
logstool would let these audits see the full daily run set, making fleet health and API-cost metrics trustworthy.Suggested Agent
MCP / gh-aw infra agent — add pagination or increase the 120s fetch window in the
logsbridge.Estimated Effort
Medium (1-4 hours)
Data Source
DeepReport Intelligence Briefing — 2026-07-20. See Workflow Audit (#46683), Safe Output Health (#46714), API Consumption (#46808), prior #45510.