You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Window evaluated: last 24 full hours (UTC), ending at this workflow's start (2026-10-26T23:xx UTC)
Total runs analyzed: 410 executed runs (out of 553 run directories collected; 143 were non-execution/obsolete-PR runs with no agent job and were excluded)
Detection-enabled runs: 406 (99.0%)
Regular runs (gh-aw-detection: false): 4 (1.0%)
Misconfigured workflows found: 0
Note
No misconfigured workflows were detected in this window. All 395 detection jobs observed (344 ran and succeeded, 51 were cleanly skipped) completed without a failing step, no workflow exceeded the >3-run threshold while explicitly disabling detection, and no workflow alternated between detection-enabled and detection-disabled runs.
Comparison Chart
Metric
Regular Runs
Detection Runs
Run count
4
406
Success rate
50.0%
91.13%
Failure count
2
36
Avg. token usage
5,432
162,985
Misconfigured Workflows
No misconfigured workflows detected in this window.
View All Run Metrics
178 distinct workflows ran in this window. Top 20 by run count:
Workflow
Detection
Runs
Successes
Avg Tokens
Engine(s)
PR Sous Chef
enabled
62
61
597
pi
Issue Monster
enabled
38
37
68,298
pi
Test Quality Sentinel
enabled
19
19
9,561
copilot
Ponytail Reviewer
enabled
18
18
87,937
codex
PR Code Quality Reviewer
enabled
17
16
9,264
copilot
Matt Pocock Skills Reviewer
enabled
17
17
35,069
copilot
Design Decision Gate 🏗️
enabled
16
15
2,802
pi
Daily Go Test Parallelizer
enabled
12
12
1,306,385
codex
Avenger
enabled
10
10
304,302
copilot
Auto-Triage Issues
enabled
8
8
118,881
codex
Contribution Check
enabled
6
6
17,801
copilot
AI Moderator
enabled
6
6
881
pi
PR Triage Agent
enabled
4
4
60,607
copilot
Code Scanning Fixer
enabled
4
4
10,999
copilot
[aw] Failure Investigator (6h)
enabled
4
4
440
claude
Smoke Claude Dynamic Workflow
enabled
3
3
1,415
claude
Smoke Copilot
disabled
3
1
2,937
copilot
Daily Credit Limit Test
enabled
2
1
38,535
pi
Release
enabled
2
2
10,920
copilot
Daily GitHub Docs SEO Optimizer
disabled
1
1
12,917
copilot
Remaining 159 workflows each had exactly 1 run in the window (predominantly schedule-triggered "Daily ..." workflows), all with detection enabled.
Note on coverage: the initial log pull hit a 3,000-run collection limit after covering only ~3.6 hours of the 24h window; a follow-up continuation fetch (using the provided before_run_id/date-range cursor) completed the remaining range, so this report reflects the full 24-full-hour window.
View Historical Trend
27 distinct days of history now recorded in cache-memory (trending/detection-metrics/history.jsonl).
Recommendations
No corrective action required this cycle — detection coverage is effectively universal (99.0% of runs) and the detection job itself has a 100% clean completion rate (success or cleanly skipped) across the window.
Smoke Copilot and Daily GitHub Docs SEO Optimizer are the only two workflows running with gh-aw-detection: false. Neither crossed the >3-run-in-window threshold used for the "frequent + disabled" misconfiguration rule, but since both are smoke/test workflows it's worth confirming the opt-out is intentional (e.g. to avoid detection noise on synthetic test content) rather than a leftover from debugging.
Smoke Copilot had a 1/3 success rate in this window (2 failures) — unrelated to detection (it ran with detection disabled throughout), but worth a separate look if it keeps failing.
Consider re-running the misconfiguration rule against a genuine 7-day window (this report's "frequent" check only had 24h of data available as a proxy) to catch workflows that disable detection intermittently over a longer horizon.
reacted with thumbs up emoji reacted with thumbs down emoji reacted with laugh emoji reacted with hooray emoji reacted with confused emoji reacted with heart emoji reacted with rocket emoji reacted with eyes emoji
Uh oh!
There was an error while loading. Please reload this page.
Summary
gh-aw-detection: false): 4 (1.0%)Note
No misconfigured workflows were detected in this window. All 395
detectionjobs observed (344 ran and succeeded, 51 were cleanly skipped) completed without a failing step, no workflow exceeded the >3-run threshold while explicitly disabling detection, and no workflow alternated between detection-enabled and detection-disabled runs.Comparison Chart
Misconfigured Workflows
No misconfigured workflows detected in this window.
View All Run Metrics
178 distinct workflows ran in this window. Top 20 by run count:
Remaining 159 workflows each had exactly 1 run in the window (predominantly
schedule-triggered "Daily ..." workflows), all with detection enabled.Note on coverage: the initial log pull hit a 3,000-run collection limit after covering only ~3.6 hours of the 24h window; a follow-up continuation fetch (using the provided
before_run_id/date-range cursor) completed the remaining range, so this report reflects the full 24-full-hour window.View Historical Trend
27 distinct days of history now recorded in cache-memory (
trending/detection-metrics/history.jsonl).Recommendations
detectionjob itself has a 100% clean completion rate (success or cleanly skipped) across the window.gh-aw-detection: false. Neither crossed the >3-run-in-window threshold used for the "frequent + disabled" misconfiguration rule, but since both are smoke/test workflows it's worth confirming the opt-out is intentional (e.g. to avoid detection noise on synthetic test content) rather than a leftover from debugging.Smoke Copilothad a 1/3 success rate in this window (2 failures) — unrelated to detection (it ran with detection disabled throughout), but worth a separate look if it keeps failing.References:
All reactions