Workflow Health — 2026-09-28
Executive read: Outcome quality is weak (51.72% acceptance), unresolved volume is concentrated in PR Sous Chef and Issue Monster, and PR Sous Chef is underdefined while most other workflows are still in flight.
Legend:
- Status: 🟩 accepted · 🟥 rejected · 🟨 pending · ⬜ unknown
- Lifecycle health: 🟢 resolving · 🟡 in flight · 🟠 aging · 🔴 stuck · ⚪ underdefined
- References: one linked item per status emoji, in the same order as the Status column
🔴 Action Items
- Highest-waste workflows — Status check workflow fixing (~18% rejected) and Issue Monster (~0% rejected but heavy unresolved volume) are the top attention workflows by waste/volume; any workflow crossing 25% waste should trigger prompt/safe-output review.
- Stuck pending items — No items are pending >48h in this cycle; no workflows classify as 🔴 stuck from available cache history.
- Underdefined workflows — PR Sous Chef, Status check workflow fixing, Contribution Check, Daily Secrets Analysis Agent, Daily Issues Report Generator, Copilot PR Prompt Pattern Analysis, and Daily Regulatory Report Generator need clearer acceptance/rejection criteria or stronger evaluators.
- Low zero-touch workflows — Zero-touch rate is 0% (0/15 accepted): all accepted items require visible human engagement, indicating output quality should be improved.
- High ignored rate — Ignored is 5/118 (4.24%), below the 30% alert threshold.
- Data quality: fallback evaluations —
fallback_exists_only_count is 0 (0%), so no weak fallback-signal inflation is detected.
Detailed metrics, evidence quality, workflow counts, and trends
Outcome Scorecard — 2026-09-28
| Metric |
Value |
Status |
| Acceptance rate |
51.72% |
🔴 <60% |
| Zero-touch rate |
0.00% |
🔴 <25% |
| Waste rate |
11.86% |
🟡 10-25% |
| Median time to resolution |
— |
— |
| Accepted |
15 / 118 |
— |
| — strong evidence |
1 |
merged, completed, approved |
| — medium evidence |
14 |
engaged, retained |
| — weak evidence |
0 |
existence only |
| Rejected |
14 |
— |
| Ignored |
5 |
no observable follow-up |
| Zero-touch |
0 / 15 |
— |
| Pending |
52 |
— |
| Runs checked |
61 |
— |
Per-Workflow Breakdown
| Workflow |
Accepted |
Rejected |
Ignored |
Pending |
Acceptance |
Zero-touch |
| Status check workflow fixing |
1 |
6 |
0 |
7 |
14.29% |
0.00% |
| Issue Monster |
3 |
0 |
0 |
37 |
100.00% |
0.00% |
| PR Sous Chef |
0 |
0 |
0 |
15 |
— |
— |
| Daily Regulatory Report Generator |
1 |
0 |
1 |
0 |
100.00% |
0.00% |
| CODEOWNERS Assistant |
0 |
0 |
0 |
2 |
— |
— |
| Review Rotation Assistant |
0 |
0 |
0 |
2 |
— |
— |
| Review Summary Assistant |
0 |
0 |
0 |
2 |
— |
— |
| Daily spec maintenance issues |
0 |
0 |
0 |
2 |
— |
— |
| Agent Persona Explorer |
0 |
0 |
0 |
1 |
— |
— |
| Contribution Check |
0 |
0 |
0 |
0 |
— |
— |
| Copilot PR Prompt Pattern Analysis |
0 |
0 |
1 |
0 |
— |
— |
| Daily Issues Report Generator |
0 |
0 |
1 |
0 |
— |
— |
| Daily Secrets Analysis Agent |
0 |
0 |
1 |
0 |
— |
— |
| Daily Spec Coverage Review |
0 |
0 |
0 |
1 |
— |
— |
| CI Insights Keeper |
1 |
0 |
0 |
0 |
100.00% |
0.00% |
| CLI UX Critic |
1 |
0 |
0 |
0 |
100.00% |
0.00% |
| Copilot PR Draft Critic |
1 |
0 |
0 |
0 |
100.00% |
0.00% |
| Daily Dependency Report Generator |
1 |
0 |
0 |
0 |
100.00% |
0.00% |
| Daily Project Performance Summary Generator (Using MCP Scripts) |
1 |
0 |
0 |
0 |
100.00% |
0.00% |
| Daily Runbook Drift Auditor |
1 |
0 |
0 |
0 |
100.00% |
0.00% |
| Issue2PR Agent |
1 |
0 |
0 |
0 |
100.00% |
0.00% |
| PR Changelog Assistant |
1 |
0 |
0 |
0 |
100.00% |
0.00% |
| PR Hype Bot |
1 |
0 |
0 |
0 |
100.00% |
0.00% |
| PR Test Plan Assistant |
1 |
0 |
0 |
0 |
100.00% |
0.00% |
| Project Lifecycle Assistant |
1 |
0 |
0 |
0 |
100.00% |
0.00% |
| Regular expression helper |
1 |
0 |
0 |
0 |
100.00% |
0.00% |
| Terminal Transcript Wrangler |
1 |
0 |
0 |
0 |
100.00% |
0.00% |
| Weekly issue metrics report |
1 |
0 |
0 |
0 |
100.00% |
0.00% |
Trend Signal
- ⬇️ Regressing: acceptance rate moved from 100.00% to 51.72% (−48.28pp), and waste rate moved from 0.00% to 11.86% (+11.86pp).
- ➡️ Stable: zero-touch rate remained at 0.00%.
📊 Measured by Outcome Collector · codex · gpt53codex · 17.4 AIC · ⌖ 4.96 AIC · ⊞ 15.9K
Workflow Health — 2026-09-28
Executive read: Outcome quality is weak (51.72% acceptance), unresolved volume is concentrated in PR Sous Chef and Issue Monster, and PR Sous Chef is underdefined while most other workflows are still in flight.
Legend:
🔴 Action Items
fallback_exists_only_countis 0 (0%), so no weak fallback-signal inflation is detected.Detailed metrics, evidence quality, workflow counts, and trends
Outcome Scorecard — 2026-09-28
Per-Workflow Breakdown
Trend Signal