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) — 2026-07-13T13:12:41Z → 2026-07-14T13:12:41Z
Total runs analyzed: 103
Detection-enabled runs: 29 (28.2%)
Regular runs: 74 (71.8%)
Misconfigured workflows found: 5
Warning
5 workflows show detection misconfigurations in this window. See the Misconfigured Workflows section for recommended fixes.
Detection-enabled runs outperformed regular runs on reliability this window (93.1% vs 82.4% success) at a comparable token cost (~38.8k vs ~36.5k tokens/run). Note the regular-run success rate is depressed almost entirely by Smoke CI (9/9 failures, 0 agent tokens — a non-agent CI smoke test); excluding it, regular agent runs succeeded at ~93.8% (61/65).
Comparison Chart
Metric
Regular Runs
Detection Runs
Total runs
74
29
Success rate (%)
82.4%
93.1%
Avg tokens
36,472
38,786
Failure count
13
2
Misconfigured count
—
5
(a name="misconfigured-workflows")(/a)
Misconfigured Workflows
Workflow
Misconfiguration type
Run count
Recommended fix
Smoke CI
Detection explicitly disabled on active workflow (>3 runs)
9
Remove gh-aw-detection: false or set true; workflow is active (9 runs) and should be monitored. Also investigate the 9/9 failure streak.
Add gh-aw-detection: true — analysis workflows should have detection enabled.
No detection-job step failures (Rule 3) and no intra-window detection state flips (Rule 4) were detected. The 2 detection-enabled failures (Copilot PR Conversation NLP Analysis, Daily Max Ai Credits Test) were general workflow failures, not detection-step failures.
Detection column:on = gh-aw-detection: true, off = explicitly false, absent = flag unset, mixed = varies within window.
View Historical Trend
18 days of history available (2026-06-27 → 2026-07-14). Detection adoption has stabilized at ~28–35 runs/day with the detection success-rate band holding in the 80–95% range.
Recommendations
Smoke CI — highest priority. It carries an explicit gh-aw-detection: false on an active (9-run) workflow and is failing every run (9/9, 0 tokens). Re-enable detection and triage the failing smoke test; it is the single largest driver of the regular-run failure count.
Enable detection on the four audit/analysis workflows (Copilot Agent Prompt Clustering Analysis, Daily Agentic Workflow AIC Usage Audit, GitHub MCP Structural Analysis, Typist - Go Type Analysis) by adding gh-aw-detection: true to their frontmatter — analysis/audit workflows benefit most from run-behavior detection.
Adoption is healthy (28.2% of runs, 93.1% success) — continue rolling detection out to the remaining absent workflows, prioritizing high-token ones (e.g. Dev, Instructions Janitor, Terminal Stylist) where behavior monitoring adds the most value.
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
2026-07-13T13:12:41Z→2026-07-14T13:12:41ZWarning
5 workflows show detection misconfigurations in this window. See the Misconfigured Workflows section for recommended fixes.
Detection-enabled runs outperformed regular runs on reliability this window (93.1% vs 82.4% success) at a comparable token cost (~38.8k vs ~36.5k tokens/run). Note the regular-run success rate is depressed almost entirely by Smoke CI (9/9 failures, 0 agent tokens — a non-agent CI smoke test); excluding it, regular agent runs succeeded at ~93.8% (61/65).
Comparison Chart
(a name="misconfigured-workflows")(/a)
Misconfigured Workflows
gh-aw-detection: falseor settrue; workflow is active (9 runs) and should be monitored. Also investigate the 9/9 failure streak.gh-aw-detection: truegh-aw-detection: true— analysis workflows should have detection enabled.gh-aw-detection: truegh-aw-detection: true— audit workflows should have detection enabled.gh-aw-detection: truegh-aw-detection: true— analysis workflows should have detection enabled.gh-aw-detection: truegh-aw-detection: true— analysis workflows should have detection enabled.No detection-job step failures (Rule 3) and no intra-window detection state flips (Rule 4) were detected. The 2 detection-enabled failures (Copilot PR Conversation NLP Analysis, Daily Max Ai Credits Test) were general workflow failures, not detection-step failures.
View All Run Metrics (45 workflows)
Detection column:
on=gh-aw-detection: true,off= explicitlyfalse,absent= flag unset,mixed= varies within window.View Historical Trend
18 days of history available (2026-06-27 → 2026-07-14). Detection adoption has stabilized at ~28–35 runs/day with the detection success-rate band holding in the 80–95% range.
Recommendations
gh-aw-detection: falseon an active (9-run) workflow and is failing every run (9/9, 0 tokens). Re-enable detection and triage the failing smoke test; it is the single largest driver of the regular-run failure count.gh-aw-detection: trueto their frontmatter — analysis/audit workflows benefit most from run-behavior detection.absentworkflows, prioritizing high-token ones (e.g. Dev, Instructions Janitor, Terminal Stylist) where behavior monitoring adds the most value.References:
Warning
Firewall blocked 1 domain
The following domain was blocked by the firewall during workflow execution:
awmgmcpgSee Network Configuration for more information.
Beta Was this translation helpful? Give feedback.
All reactions