Passed 15 of 19 test cases with the skill, and 13 of 19 without it.
Without skill68%
With skill77%
Checks it passed that the model alone missed
Lists PR URL, Title, and Target branch as bullet points
Employs $(cat <<'EOF' ... EOF) for the --body argument value
Contains the exact markdown header level 2 section ## Summary
and 2 more in the full report
3 fixed14 passed either way4 still failing1 broke
Pass-rate lift
+9 pp
Tokens
-35%
Agent turns
0%
Response time
-52%
Model gemini-3.6-flashMeasured 8/21/2026v117 of 22 cases passed with skill · 19 comparable
Tokens, turns and time describe resource use, not a dollar price.
See it in action
One of 3 test cases the skill fixed. Same model, same prompt, with and without the skill.
1 of 3
Prompt
Show the full CLI invocation to create a non-draft pull request titled 'Fix memory leak in logger' with a summary and test plan body via heredoc in standard bash.
What was checked, without and with the skill
Checked
Without skill
With skill
Uses gh pr create --title "Fix memory leak in logger"
Passed
Passed
Employs $(cat <<'EOF' ... EOF) for the --body argument value
Missed
Passed
case-12 · gemini-3.6-flash · 8/21/2026 · answers are capped and redacted
What's inside
The instructions your model reads when the skill loads.
SKILL.md59 linessingle file
Pre-check
git branch --show-current
git status
git log --oneline <main-branch>..HEAD # Check CLAUDE.md for main branch name