Description
The Copilot Session Insights workflow's orphan-branch escalation check (.github/workflows/copilot-session-insights.md:203) filters PR assignees with select(. == "copilot-swe-agent"), but the repository's actual Copilot-assignment login is "Copilot" — verified live via gh api repos/github/gh-aw/pulls, which shows assignee logins of Copilot on every currently-assigned PR (e.g. #54523, #54522, #54521), never copilot-swe-agent. The check therefore never positively matches real assignee data, so today's "0 orphaned branches" result is correct only by coincidence (no copilot/* branch had gate activity today), not because the detector actually works.
Expected Impact
Fixes a silently-broken detector so the "0%-orphan" streak (self-reported by this workflow across dozens of prior snapshots) can actually be trusted as validated evidence instead of lucky timing. Restores the workflow's core anti-abandonment signal (PRs with ≥5 simultaneous gate firings and no agent assigned for >2 hours).
Suggested Agent
Copilot coding agent — one-line fix to the assignee-match string in the jq script.
Estimated Effort
Quick (< 1 hour)
Data Source
DeepReport analysis, 2026-08-21 cycle, sourced from discussion #54464 (Daily Copilot Agent Session Analysis), "Orphaned Branch Escalation Alerts" section. Verified live via gh api in this analysis; no existing open issue found covering this specific detection-string mismatch.
Generated by 🔬 Deep Report · agent · 186.6 AIC · ⌖ 14.7 AIC · ⊞ 11.9K · ◷
Description
The Copilot Session Insights workflow's orphan-branch escalation check (
.github/workflows/copilot-session-insights.md:203) filters PR assignees withselect(. == "copilot-swe-agent"), but the repository's actual Copilot-assignment login is"Copilot"— verified live viagh api repos/github/gh-aw/pulls, which shows assignee logins ofCopiloton every currently-assigned PR (e.g. #54523, #54522, #54521), nevercopilot-swe-agent. The check therefore never positively matches real assignee data, so today's "0 orphaned branches" result is correct only by coincidence (nocopilot/*branch had gate activity today), not because the detector actually works.Expected Impact
Fixes a silently-broken detector so the "0%-orphan" streak (self-reported by this workflow across dozens of prior snapshots) can actually be trusted as validated evidence instead of lucky timing. Restores the workflow's core anti-abandonment signal (PRs with ≥5 simultaneous gate firings and no agent assigned for >2 hours).
Suggested Agent
Copilot coding agent — one-line fix to the assignee-match string in the jq script.
Estimated Effort
Quick (< 1 hour)
Data Source
DeepReport analysis, 2026-08-21 cycle, sourced from discussion #54464 (Daily Copilot Agent Session Analysis), "Orphaned Branch Escalation Alerts" section. Verified live via
gh apiin this analysis; no existing open issue found covering this specific detection-string mismatch.