[prompt-analysis] Copilot PR Prompt Analysis - 2026-10-18

[github-actions[bot]]

🤖 Copilot PR Prompt Pattern Analysis - 2026-10-18

Summary

Analysis Period: Last 30 days
Total PRs: 602 | Merged: 499 (82.9%) | Closed: 98 (16.3%) | Open: 5

Outcome rate among completed PRs: 83.6% merged. No PRs had empty bodies.

Prompt Categories and Success Rates (by title keywords)

Category	Total	Merged	Success Rate
Bug Fix	66	57	86.4%
Feature Addition	92	74	80.4%
Refactoring	13	9	69.2%
Documentation	27	22	81.5%
Testing	20	17	85.0%
Other	379	320	84.4%

Prompt Analysis

✅ Successful patterns

• Average prompt length (merged): ~161 words
• Bug fix and test prompts perform best
• Narrow, targeted tasks (e.g. version bumps, workflow tweaks)

Examples:

1. PR Expose GitHub MCP and web fetch in Smoke Codex #69747: Expose GitHub MCP and web fetch in Smoke Codex → Merged
2. PR Add fallback pricing for the Copilot AOAI API-key smoke workflow #69746: Add fallback pricing for the Copilot AOAI API-key smoke workflow → Merged
3. PR Bump MCP Gateway to v0.4.35 #69581: Bump MCP Gateway to v0.4.35 → Merged

❌ Unsuccessful patterns

• Average prompt length (closed): ~202 words (about 25% longer)
• Refactoring and broad feature/design work have lowest rates

Examples:

1. PR Fix cached audit ledger schema fixture causing CI failure #69940: Fix cached audit ledger schema fixture causing CI failure → Closed
2. PR Add opt-in duplicate PR cleanup to update and upgrade #69721: Add opt-in duplicate PR cleanup to update and upgrade → Closed
3. PR docs(specs): model routing analytics design #69507: docs(specs): model routing analytics design → Closed

Key Insights

• Length: closed-PR prompts are longer on average (202 vs 161 words), suggesting scope creep hurts.
• Refactoring is the weakest category (69%), vs 86% for bug fixes.
• Rates are stable versus yesterday (83.8% → 83.6%).

Recommendations

1. DO: Keep tasks narrowly scoped with a single clear outcome.
2. DO: Reference concrete failures/files for bug fixes and tests.
3. AVOID: Broad refactors or design-heavy multi-feature prompts; split them up.

Historical Trends

Date	PRs	Success Rate	Top Category
2026-10-18	602	83.6%	other
2026-10-17	604	83.8%	other

Trend: Stable.

---

Generated by Copilot PR Prompt Analysis (Run: 39164428075)

Generated by 🔍 Copilot PR Prompt Pattern Analysis · copilot · auto · 13.8 AIC · ⌖ 0.712 AIC · ⊞ 11.5K · ◷

• expires on Oct 19, 2026, 12:45 PM UTC-08:00