agentleFS
ClickHouseClickHouse/mcp-clickhouse/.claude/skills/release/SKILL.md

release

Skill908 starsChanged 3 days agoApache-2.0View on GitHub

Prepare and ship a new mcp-clickhouse release — bump the version, update the changelog, refresh the lockfile, then walk through tagging, PyPI, Docker, and MCP Registry publishing. Use when asked to cut a release, bump the version, or prep a release PR.

  • Commits and pushes

193 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use create-agent-adapter, changelog-style and 10 more.

Other files people use alongside this one.

Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
  • MicrosoftSkill

    create-agent-adapter

    Technical guide for creating a new Paperclip agent adapter. Use when building a new adapter package, adding support for a new AI coding tool (e.g. a new CLI agent, API-based agent, or custom process), or when modifying the adapter system. Covers the required interfaces, module structure, registration points, and conventions derived from the existing claude-local and codex-local adapters.

    Also in 5 projects that use this

  • GrinvSkill

    changelog-style

    Turn a commit range into CHANGELOG.md entries following common-changelog style. Use when finishing a PR or cutting a release, to write/fix the [Unreleased] section's entries — moving [Unreleased] under a new dated version heading is a separate, automated step (see the release skill / scripts/sync-version.ts), not this skill's job.

    Also in 4 projects that use this

  • GrinvSkill

    docs-consistency-check

    Check README/manifest.json/server.json/CHANGELOG.md/AGENTS.md and docs/*.md for drift against the actual registered tools and source. Use after adding, renaming, or removing a tool, or as part of a live-audit pass.

    Also in 4 projects that use this

  • Microsoft AzureSkill

    component-release

    Prepares a verifiable Agent Landing Zone UI release. Use for release branches, VERSION, changelog entries, tags, release notes, and compatibility evidence; never publish without explicit human approval.

    Also in 4 projects that use this

  • Microsoft AzureSkill

    documentation-consistency

    Keeps AI Landing Zone contributor, user, and operator documentation aligned with shipped Bicep behavior. Use for parameters, defaults, outputs, topology, deployment flow, operations, or releases.

    Also in 8 projects that use this

  • rivet-devSkill

    launch-graphics

    Generate agentOS marketing/launch graphics (launch & social heroes, code-snippet cards) from the committed generators in docs-internal/graphics. Use when asked to make, update, or render an agentOS launch image, social image, "agentOS Apps" graphic, or technical code-snippet image.

    Also in 3 projects that use this

  • +3 more from Grinv

The file

We split this skill into the rules it gives an agent

12 rules. 0 are common, 4 are distinctive, and 8 are very specific to this project.

  • 0

    Common

    Most of the 183 edited versions we read keep these.

  • 4

    Distinctive

    This author's own choices. Worth a look.

  • 8

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Read the current version:

    line 17
  2. Distinctive:

    If no argument was given, propose the next patch bump (increment the last segment of the current version) and confirm it with the user via …

    line 26
  3. Distinctive:

    If not, stop and ask.

    line 29
  4. Distinctive:

    Create a local release branch off the now-current main:

    line 42
  5. Very specific:

    Check CHANGELOG.md has a non-empty ## Unreleased section (at least one bullet under it).

    line 46
  6. Very specific:

    Do not touch server.json — its version fields are rewritten automatically from pyproject.toml by the MCP Registry publish workflow (Step 6) …

    line 58
  7. Very specific:

    Do not commit, push, or open a PR unless the user explicitly asks you to.

    line 66
  8. Very specific:

    Wait for it to succeed (gh run list --workflow=publish.yml) before moving on.

    line 99
  9. Very specific:

    Must run last — publish-mcp.yml verifies both the PyPI package and the GHCR image exist for this version before publishing, and it rewrites …

    line 103
  10. Very specific:

    Check the release-docker.yml run for vX.Y.Z has succeeded first (gh run list --workflow=release-docker.yml).

    line 107
  11. Very specific:

    Run publish.yml → PyPI (Step 6)

    line 120
  12. Very specific:

    Run publish-mcp.yml → MCP Registry, must be last (Step 7)

    line 121

“Common” is measured: kept by at least 6 in 10 of the 183 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 mcp-clickhouse

This project has no automated release pipeline beyond a version-bump PR — everything after that is a chain of manual GitHub Actions runs. This skill drives the whole chain, but treats anything that pushes, tags, or publishes as requiring explicit user confirmation first — those are visible/irreversible actions, not local edits.

Step 1: Determine the target version

Read the current version:

grep -m1 '^version' pyproject.toml

$ARGUMENTS is the target version hint (e.g. 0.4.1, with or without a leading v). Strip any leading v.

  • If no argument was given, propose the next patch bump (increment the last segment of the current version) and confirm it with the user via AskUserQuestion before proceeding.
  • Validate the target is a valid X.Y.Z semver and strictly greater than the current version. If not, stop and ask.

Step 2: Preflight

  • git status --short must be clean. If dirty, stop and tell the user — do not stash or discard anything automatically.
  • Switch to main and pull to make sure it's current:
    git checkout main
    git pull
  • Re-read the version from pyproject.toml now that main is current. Re-run the Step 1 default and the "strictly greater" check against it. If it changed, stop and ask.
  • Create a local release branch off the now-current main:
    git checkout -b release/X.Y.Z
  • Check CHANGELOG.md has a non-empty ## Unreleased section (at least one bullet under it). If it's empty, warn the user there may be nothing to release yet and confirm before continuing.

Step 3: Version bump edits

Make these three edits (this mirrors past release-prep commits, e.g. 5645bc9):

  1. pyproject.toml — bump version = "..." to the target version.
  2. CHANGELOG.md — rename the ## Unreleased heading to ## X.Y.Z - <date>. Get the date with date +%Y-%m-%d — never guess or hardcode it.
  3. uv.lock — run uv lock so the mcp-clickhouse package entry's version matches.

Do not touch server.json — its version fields are rewritten automatically from pyproject.toml by the MCP Registry publish workflow (Step 6) at publish time.

Step 4: Review & hand off

Stage just those three files (git add pyproject.toml CHANGELOG.md uv.lock, never git add -A), then show git diff --staged for review.

Do not commit, push, or open a PR unless the user explicitly asks you to. Leave the change staged on the release/X.Y.Z branch for their review.

Everything below only applies after the version-bump PR has merged to main.

Step 5: Tag and cut the GitHub release

Tagging and releasing is what triggers release-docker.yml, which auto-publishes multi-arch Docker images to GHCR (ghcr.io/clickhouse/mcp-clickhouse:X.Y.Z, :X.Y, :latest).

Confirm with the user before running any of this — pushing a tag and publishing a GitHub release are visible, hard-to-reverse actions.

Tag the merged commit on main, not the local release branch:

git checkout main
git pull
git tag vX.Y.Z
git push origin vX.Y.Z
gh release create vX.Y.Z --title vX.Y.Z --notes "<paste the new CHANGELOG.md section>"

Step 6: Publish to PyPI

Manual only — publish.yml has no automatic trigger.

Confirm with the user before running — this publishes an immutable artifact to PyPI.

gh workflow run publish.yml --ref vX.Y.Z

Wait for it to succeed (gh run list --workflow=publish.yml) before moving on.

Step 7: Publish to the MCP Registry

Must run last — publish-mcp.yml verifies both the PyPI package and the GHCR image exist for this version before publishing, and it rewrites server.json's version fields from pyproject.toml automatically.

Check the release-docker.yml run for vX.Y.Z has succeeded first (gh run list --workflow=release-docker.yml).

Confirm with the user before running.

gh workflow run publish-mcp.yml -f ref=vX.Y.Z

Order summary

  1. Bump version/changelog/lockfile → merge to main (Step 3–4)
  2. Tag vX.Y.Z + GitHub release → Docker image auto-publishes to GHCR (Step 5)
  3. Run publish.yml → PyPI (Step 6)
  4. Run publish-mcp.yml → MCP Registry, must be last (Step 7)

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.