[prompt-analysis] Copilot PR Prompt Analysis - 2026-11-04

[github-actions[bot]]

🤖 Copilot PR Prompt Pattern Analysis - 2026-11-04

Summary

Analysis Period: Last 30 days
Total PRs: 554 | Merged: 475 (85.7%) | Closed: 76 (13.7%) | Open: 3

No PRs had empty bodies. Categories are keyword-matched (approximate).

Prompt Categories and Success Rates

Category	Total	Merged	Closed	Success Rate
Bug Fix	141	122	19	86.5%
Feature Addition	107	90	15	85.7%
Refactoring	27	21	6	77.8%
Documentation	37	32	5	86.5%
Testing	37	33	4	89.2%
Other	205	177	27	86.8%

Prompt Analysis

✅ Successful patterns

• Avg prompt length: 176 words
• 25.7% of merged bodies reference specific files
• Examples: Add missing package specifications and unblock librarian audits #74859 Add missing package specifications and unblock librarian audits; Use a single page-level heading on the homepage #74795 Use a single page-level heading on the homepage; Refactor shared scattered-time finalization #74794 Refactor shared scattered-time finalization

❌ Unsuccessful patterns

• Avg prompt length: 145 words
• Only 13.2% of closed bodies reference specific files
• Skew toward investigation/verification tasks and CI false-positive handling
• Examples: Investigate clustered smoke-engine failures #74790 Investigate clustered smoke-engine failures; Verify reported command-injection findings are already mitigated #74788 Verify reported command-injection findings are already mitigated; Suppress runner-guard false positive in Agy installation #74751 Suppress runner-guard false positive in Agy installation

Key Insights

• File references correlate with merging (25.7% vs 13.2%).
• Longer, detailed prompts do slightly better (176 vs 145 words).
• Refactoring is weakest (77.8%, small sample of 27).
• Investigate/Verify prompts are over-represented among closures.

Recommendations

1. DO: Name target files/functions and the expected end state.
2. DO: Include context (failing run, error text, acceptance criteria).
3. AVOID: Open-ended investigate/verify prompts without a concrete deliverable.
4. AVOID: Broad refactors; split into file-scoped tasks.

Historical Trends

Date	PRs	Merged	Closed	Merge Rate
2026-11-04	554	475	76	85.7%
2026-11-03	564	479	81	84.9%
2026-10-31	600	499	91	83.2%
2026-10-30	595	490	93	82.4%

Trend: Merge rate rising over the past week (82.4% → 85.7%).

---

Generated by Copilot PR Prompt Analysis (Run: 41908481540)

Generated by 🔍 Copilot PR Prompt Pattern Analysis · copilot · auto · ~17 AIC · ⌖ ~1 AIC · ⊞ 12.4K · ◷

• expires on Nov 5, 2026, 12:43 PM UTC-08:00

[github-actions[bot]]

This discussion has been marked as outdated by Copilot PR Prompt Pattern Analysis.

A newer discussion is available at Discussion #75239.