agent toolingby microsoft
Submit Skill Fix PR
submit-skill-fix-pr
Submit a pull request with skill fixes. Validates skill structure, bumps versions, and creates a PR with a proper description. TRIGGERS: submit skill fix, create fix PR, skill fix pull request, submit PR, push skill fix
What it changes
Passed 9 of 21 test cases with the skill, and 1 of 21 without it.
Without skill5%
With skill41%
8 fixed1 passed either way13 still failing0 broke
- Pass-rate lift
- +36 pp
- Tokens
- -53%
- Agent turns
- 0%
- Response time
- -64%
Model gemini-3.6-flashMeasured 7/24/2026v19 of 22 cases passed with skill · 21 comparable
What's inside
The instructions your model reads when the skill loads.
SKILL.md19 linessingle file
Submit Skill Fix PR
Creates a pull request after committing skill fixes in microsoft/GitHub-Copilot-for-Azure.
When to Use This Skill
- You have committed skill fixes and need to submit a PR
- You need to validate skill structure before pushing
- You want to create a properly formatted fix PR
Steps
- Install NPM dependencies in the
scripts directory, if necessary.
- From the
scripts directory run npm run frontmatter and npm run references to validate the skill structure. Fix and commit any problems.
- Ensure that skill version has been bumped for any updated SKILL.md.
- Push the branch to origin and create a PR into upstream. The PR description should include:
- A brief description of the problem(s).
- A brief description of the fix(es) and how they address the problems.
- A "Fixes #<issue_number>" note. Ask the user if you don't know the issue number.
Track record
Benchmark
9 / 21
test cases passed, vs 1 / 21 without
Adoption
- Users
- 0
- Times used
- 0
Safety and source details
SkillSafety
✓ 3 of 3 passed✓Code scan15 of 15 checks clean
0 critical · 0 warning · 0 info — scanner v4 — 7/30/2026
View 15 scanner checks
live_secretremote_code_executionreverse_shelldata_exfiltrationssrf_metadataagent_snoopinganti_refusaldestructive_commandsobfuscated_unicodeunicode_homoglyphopaque_blobsuspicious_urltool_overreachtrigger_abuseprompt_injection_phrasing
✓IntentSafe
LLM judge · does the skill do what it claims
✓ContentClean
LLM judge · policy & harmful-content review
Source
Importedmicrosoft / github-copilot-for-azure
.github/skills/submit-skill-fix-pr/SKILL.md
View on GitHub ↗License MITImported 6/14/2026Version v1