[prompt-analysis] Copilot PR Prompt Analysis - 2026-10-10

[github-actions[bot]]

🤖 Copilot PR Prompt Pattern Analysis - 2026-10-10

Summary

Analysis Period: Last 30 days
Total PRs: 574 | Merged: 481 (83.8%) | Closed: 85 (14.8%) | Open: 8 (1.4%)
Merge rate among completed PRs: 85.0%. No PRs had empty bodies.

Prompt Categories and Success Rates

Categorized by title keywords; 361 PRs matched no category ("Other").

Category	Total	Merged	Success Rate (completed)
Bug Fix	84	73	88%
Feature Addition	85	66	80%
Refactoring	13	11	85%
Documentation	21	17	81%
Testing	10	8	80%
Other	361	306	86%

Prompt Analysis

✅ Successful Prompt Patterns

• Average prompt length (merged): 157 words
• Bug-fix wording ("fix", "resolve", "regression") has the highest merge rate.
• Examples:
  1. PR Restore safe-output server startup after memory schema change #67506: Restore safe-output server startup after memory schema change → Merged
  2. PR Align workflow tool allowlists with required operations #67472: Align workflow tool allowlists with required operations → Merged
  3. PR Add tracker-ID payload regressions for PR comments and issues #67441: Add tracker-ID payload regressions for PR comments and issues → Merged

❌ Unsuccessful Prompt Patterns

• Average prompt length (closed): 182 words
• Closed PRs skew toward open-ended "investigate", broad-scope or WIP tasks.
• Examples:
  1. PR Investigate deterministic installs in compiled workflows #67469: Investigate deterministic installs in compiled workflows → Closed
  2. PR Use frontmatter SHA as workflow evaluation identity #67300: Use frontmatter SHA as workflow evaluation identity → Closed
  3. PR [WIP] Restore classifier cost attribution in routed-run audits #67230: [WIP] Restore classifier cost attribution in routed-run audits → Closed

Key Insights

• Length: Closed-PR prompts are ~16% longer on average (182 vs 157 words); longer, sprawling prompts correlate with failure.
• File references: Prompts naming specific files have an 85% merge rate (83/98) vs 85% (398/468) without, so file references are not a differentiator here.
• Task type: Bug fixes outperform feature additions by ~8 points; features carry more design ambiguity.

Recommendations

1. DO: Frame tasks as concrete bug fixes/regressions with an expected behavior.
2. DO: Keep prompts focused (~150 words) with one clear deliverable.
3. AVOID: Open-ended "investigate" tasks and broad multi-part scopes.

Historical Trends

Date	PRs	Merge Rate
2026-10-10	574	85.0%
2026-10-09	575	84.7%

Trend: Stable; merge rate and volume essentially unchanged day over day.

---

Generated by Copilot PR Prompt Analysis (Run: 38084856120)

Generated by 🔍 Copilot PR Prompt Pattern Analysis · copilot · auto · 13.4 AIC · ⌖ 0.688 AIC · ⊞ 10.9K · ◷

• expires on Oct 11, 2026, 12:48 PM UTC-08:00