agentleFS
everrunseverruns/yolop/.agents/skills/release/SKILL.md

release

Skill18 starsChanged 13 days agoMITView on GitHub

Cut a new yolop release. Prepares the release PR, runs publish-readiness checks, monitors CI publish to crates.io and the everruns Homebrew tap. Use when the user asks to release, cut a version, publish, or ship to crates.io / brew.

  • 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

3 rules. 0 are common, 3 are distinctive, and 0 are very specific to this project.

  • 0

    Common

    Most of the 183 edited versions we read keep these.

  • 3

    Distinctive

    This author's own choices. Worth a look.

  • 0

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  • could be automatic: a setting could enforce it
  1. Distinctive:

    If those two counts disagree, the clone is still shallow.

    line 37
  2. Distinctive:

    Use the version the user gave.

    line 41
  3. Distinctive:

    Do not enable auto-merge: a human must click squash so a real reviewer reads the changelog.

    line 122

“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

Goal: cut a new yolop release and verify it lands on both crates.io and the everruns/homebrew-tap Homebrew tap.

knowledge/specs/release.md owns the versioning rules, the changelog format, the CI automation contract, post-release verification, hotfix and rollback policy, and the secrets table. This skill owns the procedure. For a non-release change, use /ship instead.

Releases are agent-prepared, human-merged, CI-published. The agent never tags the release and never pushes to the tap — release.yml and cli-binaries.yml do that.

0. Sync local state

Shallow clones lie: cloud sandboxes default to depth ≈ 50 and silently drop older commits from git log.

git fetch --unshallow origin main 2>/dev/null || git fetch origin main
git fetch --tags

LATEST=$(git describe --tags --abbrev=0)
git log "$LATEST"..origin/main --oneline | wc -l
gh api "repos/everruns/yolop/compare/$LATEST...main" --jq '.total_commits'

If those two counts disagree, the clone is still shallow.

1. Pick the version

Use the version the user gave. Otherwise propose one from the unreleased commit set under the spec's versioning rules, and confirm before proceeding.

2. Update the changelog

Build the commit list mechanically, then write the section in the format the spec defines (§ Changelog Format):

git log "$LATEST"..HEAD --pretty=format:'%s' --reverse \
  | grep -v '^chore(release): prepare v'

Map commits to PRs with gh pr list --state merged --base main --limit 200 when the subjects don't carry the number.

3. Bump the version

Edit version in Cargo.toml, then cargo update -p yolop to refresh the lockfile entry.

yolop-yep is versioned separately. Bump it when its API or the wire protocol changes, and update every workspace path dependency requirement that references it.

First-party extensions under extensions/ are versioned separately too. Bump one whose code changed since the last release, in both its Cargo.toml and its plugin.json, or it is published nowhere and its manifest keeps pointing at the previous tag's binaries. python3 scripts/publish_order.py lists exactly what the release publishes:

python3 scripts/publish_order.py
git log "$LATEST"..HEAD --oneline -- extensions/ crates/

tuika and tuika-codeformatters are not released from here — they ship from everruns/tuika. Bumping their version requirements is an ordinary dependency bump, and a release that needs new toolkit behavior waits on a tuika release first.

4. Verify locally

Review the commits since the previous tag for durable knowledge impact and update the affected concepts, knowledge/index.md, and knowledge/log.md. Confirm README.md and docs/ describe the released behavior. Then run the checks in AGENTS.md.

5. Verify publish-readiness

This is the step local tests can't stand in for — it exercises the cargo publish packaging boundary, missing files referenced by Cargo.toml, and version drift.

cargo publish --dry-run -p yolop-yep   # SDK; publishes before everything else
cargo search yolop --limit 1           # crates.io version must be < X.Y.Z
grep '^version' Cargo.toml
grep '"yolop"' Cargo.lock | head -1

cargo publish --dry-run fails locally for yolop and for any extension whenever a new yolop-yep version isn't on crates.io yet — expected, not a broken release. CI validates those publishes after the SDK goes live. If the SDK dry-run fails, fix the root cause and re-run; never open a release PR with a known-broken publish path.

6. Commit, push, open the PR

git add CHANGELOG.md Cargo.toml Cargo.lock
git commit -m "chore(release): prepare vX.Y.Z"
git push -u origin "$(git branch --show-current)"

Title the PR chore(release): prepare vX.Y.Z. The body carries the full changelog section, a Publish-readiness report (which dry-runs ran, what crates.io currently serves, that Cargo.toml and Cargo.lock agree), and an unchecked Post-merge verification list covering release.yml, publish.yml, cli-binaries.yml, crates.io, and the tap formula.

Do not enable auto-merge: a human must click squash so a real reviewer reads the changelog.

7. Monitor after merge

Subscribe to PR activity so workflow completions wake you, then watch:

gh run list --workflow=release.yml      --limit 1
gh run list --workflow=publish.yml      --limit 1
gh run list --workflow=cli-binaries.yml --limit 1

Green workflows are not proof, and a matrix that silently builds one target reports success just like a correct one. Run the spec's post-release verification yourself and declare shipped only when crates.io reports X.Y.Z, every bumped library and extension crate is live, each bumped extension has its <crate>-v<version> release carrying the three server archives, and the tap formula points at vX.Y.Z. crates.io publishes near-instantly; Homebrew takes minutes for the build and tap commit, so a half-verified release looks shipped when it isn't.

On failure, read the logs (gh run view <id> --log-failed) and either re-run (transient — network or registry propagation) or roll forward with a hotfix PR per the spec. Never leave a release half-shipped.

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.