agentleFS
jyjejyje/hermes-agent-helm/.claude/skills/release-flow/SKILL.md

release-flow

Skill18 starsChanged 30 days agoMITView on GitHub

Run the Changesets-based release proposal flow for this chart end-to-end via gh CLI - trigger propose-release, review the PR, adjust the version, watch CI, and merge. Use when the user asks to cut/propose/ship a release, check on a release PR, or change its version.

5 other projects use a version of this.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

1 rule. 0 are common, 0 are distinctive, and 1 is very specific to this project.

  • 0

    Common

    Most of the 5 edited versions we read keep these.

  • 0

    Distinctive

    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
  1. Very specific:

    If there are no pending Changesets, the run exits early (has-release: false) without opening or updating a PR.

    line 30

“Common” is measured: kept by at least 6 in 10 of the 5 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Release flow (propose-release via gh CLI)

This repo's release model: Changesets under .changeset/*.md are the source of the next SemVer decision, propose-release.yaml consumes all pending entries into one reviewable PR that bumps charts/hermes-agent/Chart.yaml, and merging that PR is what ships. This skill drives the whole loop with gh, without using the GitHub web UI.

See CONTRIBUTING.md for the full conceptual background (how to write a Changeset, SemVer selection, the release-note category convention); this skill is the CLI operationalization of "How to cut a release".

Steps

  1. Trigger propose-release.yaml (workflow_dispatch, no inputs; maintainer action, never runs on fork PRs). The version is always derived from the pending Changesets on main; there is no way to pin a version or force a bump level at trigger time:

    gh workflow run propose-release.yaml

    If there are no pending Changesets, the run exits early (has-release: false) without opening or updating a PR.

  2. Watch the run:

    gh run list --workflow=propose-release.yaml --limit 1
    gh run watch <run-id>
  3. Review the release PR (branch changeset-release/main, managed by the changesets/action GitHub Action - idempotent, re-running step 1 updates the same PR):

    gh pr list --head changeset-release/main
    gh pr view <PR-number>     # combined version + changelog entries
    gh pr diff <PR-number>     # actual Chart.yaml / CHANGELOG.md diff
  4. Adjust the version if needed. There is no PR-comment command for this (no bot listens for one). Edit or add a .changeset/*.md file on main with the correct major/minor/patch impact instead of editing the generated Chart.yaml version directly, then re-run step 1 to refresh the same PR against the corrected Changesets.
  5. Check CI (validate-chart.yaml runs automatically on PR push):

    gh pr checks <PR-number> --watch
  6. Merge: this triggers release-chart.yaml, which tags vX.Y.Z, writes the GitHub Release, and publishes the chart to both the OCI registry (oci://ghcr.io/<owner>/hermes-agent-helm/hermes-agent) and the GitHub Pages Helm repo (https://<owner>.github.io/hermes-agent-helm). Use --merge (not --squash) so changesets/action's own commit structure survives:

    gh pr merge <PR-number> --merge
  7. Watch the release run, then confirm verify-release.yaml (triggered automatically afterward) also goes green - it installs the just-published OCI artifact into a fresh kind cluster and re-verifies it:

    gh run list --workflow=release-chart.yaml --limit 1
    gh run watch <run-id>
    gh run list --workflow=verify-release.yaml --limit 1

Notes

  • main currently has no branch protection: merging is immediate, no required reviews/checks are enforced by GitHub itself. Still wait for gh pr checks to go green before merging.
  • release-chart.yaml is a no-op (tag-existence guard) if the vX.Y.Z tag already exists, so re-running things after a merge is safe.
  • changeset-release/main is managed entirely by changesets/action itself; nothing in this repo's own workflows deletes or touches that branch.

Download now (ZIP)

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
What did you use it for? e.g. Reviewing pull requests in a Rails app

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.