Add GitHub Actions and Codespaces Efficiency skill#1646
Conversation
🔍 Skill Validator Results✅ All checks passed
Summary
Full validator output```text Found 2 skill(s) [github-actions-efficiency] 📊 github-actions-efficiency: 866 BPE tokens [chars/4: 1,051] (detailed ✓), 10 sections, 1 code blocks [github-codespaces-efficiency] 📊 github-codespaces-efficiency: 734 BPE tokens [chars/4: 904] (detailed ✓), 10 sections, 1 code blocks ✅ All checks passed (2 skill(s)) ``` |
There was a problem hiding this comment.
Pull request overview
Adds a new skill (github-actions-codespaces-efficiency) to guide audits and concrete improvements that reduce GitHub Actions runtime/cost and Codespaces waste, and registers it in the generated skills documentation.
Changes:
- Added a new skill entrypoint (
SKILL.md) with a focused workflow, output shape, and review rubric for Actions + Codespaces efficiency work. - Added supporting reference docs for Actions audits, Codespaces optimization, reporting, and reusable implementation patterns.
- Added agent metadata and updated
docs/README.skills.mdto include the new skill and its bundled assets.
Reviewed changes
Copilot reviewed 7 out of 7 changed files in this pull request and generated 1 comment.
Show a summary per file
| File | Description |
|---|---|
| skills/github-actions-codespaces-efficiency/SKILL.md | New skill entrypoint defining scope, workflow, output contract, and references. |
| skills/github-actions-codespaces-efficiency/references/actions.md | Reference guidance for auditing and improving GitHub Actions efficiency. |
| skills/github-actions-codespaces-efficiency/references/codespaces.md | Reference guidance for devcontainer/Codespaces sizing, prebuilds, and cost controls. |
| skills/github-actions-codespaces-efficiency/references/patterns.md | Concrete YAML/config patterns for caching, concurrency, trigger scoping, etc. |
| skills/github-actions-codespaces-efficiency/references/reporting.md | Guidance for before/after reporting and follow-up review passes. |
| skills/github-actions-codespaces-efficiency/agents/openai.yaml | Agent UI/prompt metadata for the skill. |
| docs/README.skills.md | Registers the new skill in the generated skills index with bundled assets. |
Co-authored-by: Copilot Autofix powered by AI <175728472+Copilot@users.noreply.github.com>
|
Added a compact audit command block to SKILL.md to address the review comment about having no code blocks. This is included in commit cf55938. |
aaronpowell
left a comment
There was a problem hiding this comment.
I think this would make more sense to exist as two separate skills, one that focuses on GitHub Actions and another that focuses on Codespaces/Devcontainers. While yes, it's likely people are using them together, they aren't really overlapping products in how they work, so having them in a single skill will overload the context window with (arguably) unhelpful tokens.
I did look at its context use which is why I layered the specifics for workflows, codespaces and the reporting into reference's, happy to take another look and split it if you think its better, will drop the openai file |
|
Sorry, this PR is looking incredibly confused now, it's adding an external plugin which isn't following the process for submitting external plugins and it's deleted an MCP server we use. Can you clean it up to be focused and then I'll review again. |
Pull Request Checklist
npm startand verified thatREADME.mdis up to date.stagedbranch for this pull request.Description
Adds a new
github-actions-codespaces-efficiencyskill focused on reducing GitHub Actions and Codespaces waste.The skill provides:
SKILL.mdentrypoint with scope, workflow, output shape, and review rubric.agents/openai.yamlmetadata aligned to the full Actions + Codespaces scope.npm run build.Validation run locally:
node ./eng/validate-skills.mjsnpm run buildType of Contribution
Additional Notes
This PR targets
stagedand is opened from the contributor fork branchmutl3y:workflow_cost.By submitting this pull request, I confirm that my contribution abides by the Code of Conduct and will be licensed under the MIT License.