release-version-update
Skill17 starsChanged 6 days agoMITView on GitHub
Run a user-requested fleet-harness canary synchronization and Stable Release, or continue watching that release run. Not for ordinary PR merges or editing local version numbers.
In this repository13 other files it gives its agents.
- .changelog.d/CLAUDE.mdCLAUDE.md
- CLAUDE.mdCLAUDE.md
- ai-gateway-loop-optimizationSkill
- clean-codeSkill
- console-e2eSkill
- console-handoffSkill
- console-performanceSkill
- design-sweepSkill
- git-worktreeSkill
- learning-harvestSkill
- product-proposalSkill
- pr-workflowSkill
- rebase-on-canarySkill
The file
We split this skill into the rules it gives an agent
2 rules. About 0 are common, 1 is distinctive, and 1 is very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈1
Distinctive · estimate
This author's own choices. Worth a look.
1
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 14
No required inputs. If the user requests synchronization only, stop before deployment.
- Very specific: line 23
Start from the absolute path of the actual canary checkout.
≈ “Common” is an estimate: at least 1 in 5 of the 350,495 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
Release Version Update
Execute the requested release lifecycle. That request includes direct canary work and a validated fast-forward push from origin/canary to main. Reading this document or automatic skill selection alone does not authorize deployment.
.github/workflows/stable-release.yml owns version calculation, workspace version synchronization, changelogs, release commit/tag, GitHub Release, Console release assets, and app assets. Do not edit local versions/generated changelogs or substitute local tags or asset uploads for CI. Fleet Console is no longer published to npm; never publish it there manually. Console installs and self-updates read the latest GitHub Release, so never manually publish or mark as latest a Release that lacks the Console assets.
Inputs and routes
No required inputs. If the user requests synchronization only, stop before deployment. To resume a named run, verify its identity/SHA and enter the watch phase. A version number is not an input; resolve requests conflicting with CI version calculation through the user.
| Phase | Read before execution |
| Canary checkout, fetch/synchronization, release inputs, full local checks | Prepare |
| Main FF, exact Stable Release run, canary synchronization/report | Publish and sync |
Execution contract
- Start from the absolute path of the actual
canarycheckout. Preserve user changes and stop on a dirty tree. Never send a topic worktree's HEAD tomain. - Fetch
origin/canary,origin/main, and tags. Fast-forward local canary. If public canary and main diverge, merge main into canary; never rebase/force-push. Preserve release-owned history/version and canary's unconsumed fragments/unique product work. - For an empty range, verify the published tag/Release and finish. For docs-only/ignored ranges, confirm with
release-tip-guardand report that deployment will not trigger. Do not manufacture a release with dummy files or empty commits. - For real release inputs, run full
pnpm build,pnpm typecheck,pnpm test, and fragment--check --allow-emptyfrom canary. Full verification is a release gate. Fix failures narrowly on canary, commit/push, then repeat all four checks. Do not waive local verification because CI also runs. - Recheck that current canary matches the verified SHA. If canary advanced, synchronize and verify the new tip. Push
origin/canary:mainonly when clean, remote-matching, validated, and descended from main. - Watch the Stable Release run matching the pushed SHA to completion. On failure, report job/run URL and stop. Failed-job retries are allowed; local tags/publishing or a new empty main commit are not substitutes.
- Incorporate the successful release commit into canary by FF or a merge preserving concurrent canary work. Never force-push.
Completion and reporting
Deployment is complete only after run success, required publish jobs, exactly one successful desktop build/carry path, release commit/tag, non-draft latest GitHub Release carrying the Console assets whose manifest names that version. Verify canary contains that release commit too. Running jobs are not success.
Report canary path, old/new SHAs, shipped range, consumed fragments/fragmentless commits, validation/fix SHAs, trigger gate, main pushed SHA/run URL, version/tag/Release URL/Console assets/desktop outcome, and final ref relationships. Distinguish synchronization-only, nothing-to-ship, and failure from deployment success. The host synthesizes directly from verified git/CI evidence.
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.

