n8mus/AetherOrion / .claude/skills/release-prep
release-prep
VerifiedThe AetherSDR release-prep pass — takes a CalVer version and drives the release from "main is where we want to cut" to "the prep PR is merged on main and ready to tag". Maps the range through the commit-to-PR API, writes the CHANGELOG section from PR bodies, refreshes README and ROADMAP, edits the six release files, validates, opens the prep PR, folds in what merges after the cutoff, and hands the merged squash commit to /tag-release for the tag, the release and the asset verification. Every judgment call goes to the maintainer through AskUserQuestion. Use when the user says "/release-prep 26.9.4", "/release-prep v26.9.4", "prep the release", "release prep", "write the changelog for 26.9.4", or otherwise wants a version prepped; the cut itself is "/tag-release".
| File | Kind | Size |
|---|---|---|
| SKILL.md | skill md | 29.7 KB |
| agents/openai.yaml | config | 243 B |
| references/changelog-style.md | markdown | 8.1 KB |
| references/pr-body-template.md | markdown | 5.4 KB |
| scripts/check_release_prep.py | script | 22.0 KB |
| scripts/release_contents.py | script | 19.8 KB |
references/pr-body-template.md
View on GitHub# Prep PR body — the skeleton
Title: `release: prep vX.Y.Z — <short description>` (or `docs(release): prep vX.Y.Z`).
Both forms are in the record; `release: prep` is the more common.
The repo squash-merges with the body as the commit message, so this text
becomes permanent history on `main`. It states the current state of the
branch, not the churn that produced it. Keep every number in it computed from
the mapping, and re-check them before the merge — v26.8.1's body drifted
116 → 117 → 118 while `main` moved under it.
```markdown
## Summary
Six-file release prep for **vX.Y.Z**, metadata and docs only — no source, test or CI changes.
Cutoff: `origin/main` at `<sha>` (#NNNN, merged <date>). Anything that merges after it is folded in before the tag (see "Folded in after the cutoff" below if that happened).
**N merged changes** from **M human contributors**, AetherClaude (A) and Dependabot (K). First-time contributors: **@a**, **@b** — or "none this cycle" — confirmed by looking for a prior merged PR per login before the previous tag, not from `author_association`.
## Changelog
Built from `v<prev>..<sha>` mapped through the commit-to-PR API (`repos/…/commits/<sha>/pulls`), not from commit subjects: <N> PRs and <J> direct commits, every one cited in the section. The notes lead with <headline features>. Numbers in the section come from the PR bodies; where a body disagreed with itself (#NNNN: x vs y) the section uses <x> because <reason>.
<If the previous tag was off-main:> `v<prev>` is not an ancestor of `main`; its prep landed as the squashed duplicate `<sha>` (#NNNN) and is excluded from the range.
## Docs refresh, not just version strings
- **README** — <highlights lines changed (engine count, overlay list, mode list)>, <Supported Hardware → Other radio families paragraphs touched, and why>, <the Roadmap section lines that moved>.
- **ROADMAP** — In flight: <entries advanced with a vX.Y.Z mention; entries removed as shipped>. Queued: <entries deleted as shipped>. Open RFCs: <checked against `label:rfc`; joined / left>. Recently shipped: <J> new entries carry a `(vX.Y.Z)` marker<; month-rollover trim if any>.
## Judgment calls flagged for the maintainer
- <Each decision put through the decision gate and the answer, e.g. "AetherTX (#5819) is treated as delivering the "TX DSP chain visual rebuild" roadmap item, so that line leaves In flight in both README and ROADMAP.">
- <A radio family's README maturity label, if one changed, and who decided.>
- "None — nothing in this prep needed a call." if so.
<Only when the maintainer chose to include an unmerged PR:>
> ### ⚠️ This PR documents #NNNN as shipped, and #NNNN has not merged yet
> Per maintainer direction, **#NNNN (<title>)** lands before the tag, so it is written into the CHANGELOG, `ROADMAP.md` and `README.md` here, and the stated count (N) assumes it merges as one squash commit. **Merge #NNNN first.** **If #NNNN slips, edit before tagging:** drop <the bullet at CHANGELOG § …>, <the sentence at ROADMAP § …>, <the clause at README § …>, set **@login** back to <n> commits, and set the count back to <N−1>.
## Scope
| file | change |
|---|---|
| `CMakeLists.txt` | `project(AetherSDR VERSION X.Y.Z)` |
| `README.md` | Current version line, plus the prose refresh above |
| `AGENTS.md` | Current version line |
| `CHANGELOG.md` | new `[vX.Y.Z] — YYYY-MM-DD` section; `[Unreleased]` stays and stays empty; history below byte-identical |
| `packaging/linux/io.github.aethersdr.aethersdr.metainfo.xml` | `<release version="X.Y.Z" date="YYYY-MM-DD"/>` first in `<releases>` |
| `ROADMAP.md` | cycle heading → post-vX.Y.Z, plus the prose refresh above |
| <`extra/file.md`> | <why — e.g. "the README split moved build detail here (#5673 precedent)"> |
## Validation
Output of `python .claude/skills/release-prep/scripts/check_release_prep.py --mapping mapping.json`, with each line's result:
- All six version spots agree on X.Y.Z (grepped, not trusted).
- `appstreamcli validate` — <result> (or `xmllint --noout` / XML parse, and which).
- Every PR in `v<prev>..<sha>` is cited; the only other numbers cited are <#NNNN (issue), #NNNN (RFC), #NNNN (prior PR)>.
- Intro count, Contributors counts and the first-time line re-derived from the mapping — match.
- `[Unreleased]` empty; CHANGELOG history below `[v<prev>]` byte-identical to `origin/main`.
- ROADMAP `(vX)` markers newest-first, no gap; every local link in README and ROADMAP resolves; no heading removed from README/ROADMAP is still linked elsewhere; code fences balance; no duplicated table headers or doubled `---`.
- Outside `CHANGELOG.md`, the only `<prev>` lines removed are the four current-version lines.
- `git diff --check` clean; the diff touches no `src/`, `tests/`, `third_party/` or `.github/workflows/` path.
- <Configure, if run:> `cmake` configure reports `CMAKE_PROJECT_VERSION=X.Y.Z`. No build or test run — this is metadata; CI builds it. <Never claim a build or tests that did not run.>
- `main` CI at the cutoff: <green / which job is red and why; e.g. "hl2_state_restore_test fails on the unchanged cutoff source (#NNNN); this prep does not fix it and the release ships with it red">.
<If `main` moved during review:>
## Folded in after the cutoff
- #NNNN (`<sha>`, merged <date>) — added to § <section>; counts moved N−1 → N, **@login** n−1 → n.
After merge: tag the resulting **main** commit, not this branch.
🤖 Generated with [Claude Code](https://claude.com/claude-code)
```
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| rea-changelog-updatemorluto/reaPrepare or rewrite REA release changelogs and GitHub release notes from pinned Git history, with verified contributor thanks and Release Please synchronization. | Stars: 472k | Updated | Claude CodeCodex | Verified | Install |
| academy-guideanthropics/skillsStop and check this skill before finishing any reply to a question about how to use Claude or a Claude product — it recommends matching courses, tutorials, and use cases from Claude Academy (academy.claude.com), Anthropic's learning hub. Trigger on: "how do I", "how can I", "getting started with", "what can Claude do", "teach me", "learn to use"; questions about artifacts, projects, skills, plugins, connectors, MCP; requests about rolling Claude out to a team, class, or organization; and any ask for training materials, onboarding content, or learning resources. Use it when the user is learning how to use a feature or product — not when they are mid-task and just want the task done. This skill composes with other skills: after consulting product documentation to answer how a Claude feature works, also check here for a matching course or tutorial — a docs-grounded answer and an Academy recommendation belong together. Only recommend on a strong match; never invent Academy content. | Stars: 198k | Updated | Claude CodeCodex | Verified | Install |
| claude-apianthropics/skillsReference for the Claude API / Anthropic SDK — model ids, pricing, params, streaming, tool use, MCP, agents, caching, token counting, model migration. TRIGGER — read BEFORE opening the target file; don't skip because it "looks like a one-liner" — whenever: the prompt names Claude/Anthropic in any form (Claude, Anthropic, Fable, Opus, Sonnet, Haiku, `anthropic`, `@anthropic-ai`, `claude-*`, `us.anthropic.*`, `[1m]`); the user asks about an LLM (pricing/model choice/limits/caching) — never answer from memory; OR the task is LLM-shaped with provider unstated (agent/MCP/tool-definition/multi-agent/RAG/LLM-judge/computer-use; generate/summarize/extract/classify/rewrite/converse over NL; debugging refusals/cutoffs/streaming/tool-calls/tokens). SKIP only when another provider is being worked on (overrides all triggers): OpenAI/GPT/Gemini/Llama/Mistral/Cohere/Ollama named in the query; OR `grep -rE 'openai|langchain_openai|google.generativeai|genai|mistralai|cohere|ollama'` over the project hits (run this grep FIRST if no provider named — don't Read the file). | Stars: 198k | Updated | Claude CodeCodex | Unverified | Install |
| doc-coauthoringanthropics/skillsGuide users through a structured workflow for co-authoring documentation. Use when user wants to write documentation, proposals, technical specs, decision docs, or similar structured content. This workflow helps users efficiently transfer context, refine content through iteration, and verify the doc works for readers. Trigger when user mentions writing docs, creating proposals, drafting specs, or similar documentation tasks. | Stars: 198k | Updated | Claude CodeCodex | Verified | Install |