jackin-release
Use only when the user explicitly requests this skill. Cuts a jackin❯ release — runs the readiness gates, writes the changelog, recommends a version, then executes cargo release on explicit operator confirmation.
Works with
--- name: jackin-release description: Use only when the user explicitly requests this skill. Cuts a jackin❯ release — runs the readiness gates, writes the changelog, recommends a version, then executes cargo release on explicit operator confirmation. license: Apache-2.0 --- # jackin-release **Cut** a release. The only irreversible skill in the suite — `cargo release --execute` bumps the version, rewrites the changelog header, tags, and pushes. Hard-gated on one explicit operator "yes"; nothing releases on assumption. Orchestrates `jackin-release-check` (the gates) and `jackin-release-notes` (the changelog), then recommends a version from the changelog shape and cuts. ## When to use - The operator asks to cut / ship / release a version. ## When NOT to use - Just checking readiness → `jackin-release-check`. - Just updating the changelog → `jackin-release-notes`. ## Prerequisites `cargo-release` installed; `gh` authenticated; `release.toml` with `pre-release-replacements` for `CHANGELOG.md`; `CHANGELOG.md` with the `<!-- next-header -->` marker; `.github/workflows/ci.yml`. ## Process 1. **Gates.** Run `jackin-release-check`. A red gate **stops** — show the report and name what to fix. Warns surface; ask whether to continue. *Done when* every gate is green, or the operator accepted the warns. 2. **Changelog.** Run `jackin-release-notes`. Present the `[Unreleased]` section; let the operator re-classify or reword. *Done when* the operator approves the changelog. 3. **Recommend a version.** Read the approved `[Unreleased]` buckets: any **Removed** or "breaking" entry → **major**; any **Added** → **minor**; only **Fixed** / **Changed** / **Security** / **Deprecated** → **patch**. Read the current version (`grep '^version' Cargo.toml`), compute the next, present the recommendation, wait. *Done when* the operator confirms the bump level. 4. **Commit the changelog.** If `CHANGELOG.md` has uncommitted edits, `git add CHANGELOG.md && git commit -m "docs: update changelog for vX.Y.Z"`. 5. **Final confirm.** Show the summary — version, bump level, changelog shape, and the exact `cargo release <level> --execute` about to run (it bumps `Cargo.toml`, renames `[Unreleased]` to `[X.Y.Z] - <date>` via `release.toml`, tags, and pushes). **Do not run it without an explicit "yes."** *Done when* the operator says "yes." 6. **Cut.** `cargo release <level> --execute`. On failure, show the error and stop. 7. **Verify.** `git tag -l "vX.Y.Z"` and `git ls-remote --tags origin "refs/tags/vX.Y.Z"`. Point the operator at the release workflow that now builds the artifacts. ## Error recovery - Before `cargo release`: safe to fix and re-run — this skill re-validates everything. - During `cargo release`: check what committed or tagged. Tag created but unpushed → `git push origin vX.Y.Z`. Version commit without tag → reset and retry. - After: the release is done; a CI failure is a workflow issue, not a rollback. ## Common mistakes - Running `cargo release` without the explicit "yes" — the one hard gate. - Letting a red readiness gate through to the cut. - Hand-editing the `[X.Y.Z]` version heading — `cargo-release` rewrites it via `release.toml` `pre-release-replacements`; double-editing corrupts the release commit.
More Writing & Documentation skills
paper-context-resolver
lllllllama/rigorpilot-skills
Rigor Paper Context helper for README-first deep learning repo reproduction. Use only when the README and repository files leave a narrow reproduction-critical gap and the task is to resolve a specific paper detail such as dataset split, preprocessing, evaluation protocol, checkpoint mapping, or runtime assumption from primary paper sources while recording conflicts. Do not use for general paper summary, repo scanning, environment setup, command execution, title-only paper lookup, or replacing README guidance by default.
repo-intake-and-plan
lllllllama/rigorpilot-skills
Rigor Intake helper for README-first deep learning repo reproduction. Use when the task is specifically to scan a repository, read the README and common project files, extract documented commands, classify inference, evaluation, and training candidates, and return the smallest trustworthy reproduction plan to the main orchestrator. Do not use for environment setup, asset download, command execution, final reporting, paper lookup, or end-to-end orchestration.
minimal-run-and-audit
lllllllama/rigorpilot-skills
Rigor Run skill for README-first deep learning repo reproduction. Use when the task is specifically to capture or normalize evidence from the selected smoke test or documented inference or evaluation command and write standardized `repro_outputs/` files, including patch notes when repository files changed. Do not use for training execution, initial repo intake, generic environment setup, paper lookup, target selection, hidden scientific-meaning changes, or end-to-end orchestration by itself.

