usage-monitor-release
Skill19 starsChanged 50 days agoMITView on GitHub
Cut and publish a stable usage-monitor release: pick the version, verify locally, create the annotated tag and follow the GitHub Actions workflow that publishes Windows, Linux and macOS artifacts. Use to cut, ship, publish or troubleshoot a release.
- Commits and pushes
In this repository13 other files it gives its agents.
- AGENTS.mdAGENTS.md
- CLAUDE.mdCLAUDE.md
- orchestrateSkill
- usage-monitor-commit-pushSkill
- usage-monitor-designSkill
- usage-monitor-nsis-installerSkill
- usage-monitor-release-betaSkill
- usage-monitor-release-videoSkill
- usage-monitor-social-videoSkill
- usage-monitor-test-reportSkill
- usage-monitor-token-cleanupSkill
- usage-monitor-visual-optionsSkill
- usage-monitor-workflowSkill
The file
Show its rules and text
We split this skill into the rules it gives an agent
13 rules. About 0 are common, 6 are distinctive, and 7 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈6
Distinctive · estimate
This author's own choices. Worth a look.
7
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 12
Read AGENTS.md, README.md, build.gradle.kts, and .github/workflows/release-linux.yml first.
- Very specific: line 18
Verify that main is clean and matches origin/main — the tag must point to a commit that is already on the remote main, or verify-version …
- Distinctive: line 36
Keep release-related packaging aligned:
- Distinctive: line 41
Run the local verification that must pass before the tag exists:
- Very specific: line 43
Do not package locally as a release gate.
- Very specific: line 54
create the annotated tag with the temporary agent git identity: git -c user.name=claude -c user.email=claude@anthropic.com tag -a vX.Y.Z -m …
- Distinctive: line 68
Do not release from a dirty tree unless the user explicitly wants that risk.
- Distinctive: line 69
Do not ask the user to confirm the tag or the push.
- Very specific: line 70
Do not forget that release artifacts are published by CI from v* tags.
- Distinctive: line 71
Do not reintroduce local packaging as a release gate.
- Distinctive: line 72
Do not forget that the GitHub Release body should show the commit summary for the new version.
- Very specific: line 73
Do not treat a macOS-only job failure as acceptable: the build-macos matrix blocks publish-release, and a green release without both DMGs …
- Very specific: line 74
Prefer current code and workflow files over older notes in .agents, .claude, or docs/research.md when they disagree.
≈ “Common” is an estimate: at least 1 in 5 of the 372,429 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Usage Monitor Release
Release this repository in a way that stays aligned with the current build, installer, and GitHub Actions workflow.
Workflow
- Read AGENTS.md, README.md, build.gradle.kts, and .github/workflows/release-linux.yml first.
- Invoking this skill already means "cut and publish the release". Do not ask for confirmation to commit, tag, or push — go through the whole flow, including step 8, and only report the result.
- Confirm the release type from the request:
patchminormajor
- Verify that
mainis clean and matchesorigin/main— the tag must point to a commit that is already on the remotemain, orverify-versionrefuses the release:git fetch origin && git status --short --branchgit log --oneline -5
- Pick the version. There is no version bump commit (#344): the version comes from the tag.
- Current version:
git describe --tags --abbrev=0 --match "v[0-9]*" --exclude "*-beta*"; compute the nextX.Y.Zfrom the release type. Beta tags (vX.Y.Z-beta.N, issue #355) are never the base: without--excludea beta would become the "current" stable version. If an open beta seriesvX.Y.Z-beta.Nexists, releasing it as stable means tagging that sameX.Y.Z. Beta releases have their own skill,usage-monitor-release-beta. - Open beta series? (
git -c versionsort.suffix=- tag --sort=-version:refname --list "v*-beta.*" | head -n 1with no stable of the sameX.Y.Z). Promoting it isusage-monitor-release-beta promote. A stable lower than the series (apatchof the previous stable whileX.Y.Z-beta.Nis open) is fine for everyone else but never reaches the testers —X.Y.Z-beta.Nis higher. Say so in the report, and if the fix matters to them, follow it with anextbeta. build.gradle.ktsresolvesversionfrom-PappVersion(the release workflow passes the tag) and, without it, from that samegit describe— never edit a version literal there or insrc/installer/UsageMonitor.nsi(its!ifndefdefault only serves directmakensisruns).
- Current version:
- Keep release-related packaging aligned:
- The Gradle
buildNsisInstallertask passes/DPRODUCT_VERSION=$appVersionto NSIS. - The local helper build-with-icon.ps1 is Windows-only and exists for icon-patched local packaging.
- The official GitHub release artifacts come from the tag-driven workflow in .github/workflows/release-linux.yml, which builds Linux, Windows, and macOS.
- macOS
.dmgfiles come from thebuild-macosjob (macos-latestfor arm64,macos-15-intelfor x64). They are unsigned: no Apple Developer ID, no notarization.packageDmgcannot run on Windows or Linux, so there is no local validation path for it.
- The Gradle
- Run the local verification that must pass before the tag exists:
gradlew.bat allTests— mandatory. The tag is what triggers the release; a broken suite discovered after the push means a burned tag on the remote.- Do not package locally as a release gate.
build-windowsalready runspackageInstaller, andbuild-linux/build-macoscover the other two OSes. Local packaging duplicates CI and proves only one platform. gradlew.bat packageInstalleronly when the task is debugging the installer itself (see theusage-monitor-nsis-installerskill).- Window hosts changed? Stop before the tag (#342). Run
git diff --name-only $(git describe --tags --abbrev=0)..HEADand look forHudWindow.kt,DesktopWindowFrame.kt,Main.ktor any*WindowHost*.kt. If one changed, no test covers what broke in #340 — the window system. List the rows ofdocs/hud-notch.md› "Fora do alcance dos testes" that the change touches and ask the maintainer whether the build was opened on Linux (X11). This is the only confirmation this skill asks for: without it, v41.0.0 shipped a HUD that was broken on Linux for three releases. If nothing in the list changed, continue without asking.
- Publish. Invoking this skill is the request to publish — never stop to ask for confirmation:
- create the annotated tag with the temporary agent git identity:
git -c user.name=claude -c user.email=claude@anthropic.com tag -a vX.Y.Z -m "vX.Y.Z"(annotated is mandatory —verify-versionrejects lightweight tags) - push only the tag:
git push origin vX.Y.Z. Nothing is pushed tomain, so a release no longer triggersCIorCodeQLthere.
- create the annotated tag with the temporary agent git identity:
- Watch the GitHub Actions workflow
Release Desktop Packagesand report the outcome.verify-versionchecks that the tag isvX.Y.Z, annotated and onmain;verifyrunsallTestsand the installer scenarios in parallel with the builds;publish-releaserequires both. To rebuild an existing tag without publishing, dispatch the workflow withpublish: false. The published release must carry every artifact family:- Windows:
UsageMonitor-Setup-X.Y.Z.exe, the only Windows artifact. The.msiwas dropped after v37 — both installers wrote to the same%LOCALAPPDATA%\Usage Monitorand an MSI install could never update itself. - Linux:
.deb,.rpm, andusage-monitor_X.Y.Z_linux_x64.tar.gz - macOS:
usage-monitor_X.Y.Z_macos_arm64.dmgandusage-monitor_X.Y.Z_macos_x64.dmgA release missing either DMG is incomplete and must be reported as such.
- Windows:
- The published GitHub Release should include a short changelog built from commits between the previous
v*tag and the new tag. That changelog is generated by the workflow'sGenerate release notesstep; do not create it manually withgh release create.
Guardrails
- Do not release from a dirty tree unless the user explicitly wants that risk.
- Do not ask the user to confirm the tag or the push. The skill invocation is the authorization; asking again is unwanted friction. The one exception is step 7: a window host changed since the last tag (#342).
- Do not forget that release artifacts are published by CI from
v*tags. - Do not reintroduce local packaging as a release gate; CI owns packaging for all three OSes.
- Do not forget that the GitHub Release body should show the commit summary for the new version.
- Do not treat a macOS-only job failure as acceptable: the
build-macosmatrix blockspublish-release, and a green release without both DMGs means the workflow was edited or skipped. - Prefer current code and workflow files over older notes in
.agents,.claude, ordocs/research.mdwhen they disagree.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

