Skip to content

Pixel-CLI/pixel / .agents/skills/release

release

Verified

Cut a pixel release end to end — pick the version, prepare the release commit (changelog cut, lockstep version bump, Cargo.lock, check-release) with prepare.sh in a pull request into main, tag vX.Y.Z on its merge to trigger .github/workflows/release.yml, and verify the published release, the install.sh asset and the Homebrew tap with evidence. Also covers resuming an interrupted release, patch releases while main is not releasable, and recovering from a failed Release run. Use when asked to release, publish, tag or ship a new pixel version, bump the version, cut the changelog, check a published release, or when the Release workflow failed.

FileKindSize
SKILL.mdskill md30.6 KB
check-candidate.pyscript3.2 KB
prepare.shexecutablescript23.7 KB
references/security-release.mdmarkdown3.6 KB

references/security-release.md

View on GitHub
# Private advisory release

Read this before promising a patched version or importing an advisory fix.
Keep the release record and unannounced details in the private advisory or a
private scratchpad. A public draft PR is public; delaying its merge does not
keep its patch or threat model under embargo.

1. **Choose the candidate and version first.** Apply the skill's version
   convention against the last shipped tag, including removed commands and
   flags without aliases. Decide a maintenance branch before promising a
   patch. Leave `patched_versions` unset until the choice is settled; record
   the reason and update it if the candidate changes.
2. **Validate the private patch before import.** Record base/head SHAs,
   regression tests (every consuming path, including read-only opens), full
   gates and a mutation verdict for that diff on authorized private
   infrastructure. GitHub Actions do not run in temporary advisory forks;
   waiting for their public required checks cannot produce this evidence.
   Do not push the patch publicly to obtain checks. The public
   `mutants.yml` dispatch accepts an ancestor of the dispatch branch; it is
   a post-import verification, not private pre-merge validation. If private
   infrastructure is unavailable, record the missing verdict and obtain a
   maintainer decision before import; never describe it as passed.
3. **Plan the merge with the maintainer.** Identify author, eligible reviewer
   and merge operator. Inspect the actual advisory merge UI and applicable
   rules before waiting on `Expected` checks. GitHub documents branch
   protection exceptions for advisory merges, but v0.7.0 still encountered
   a ruleset block: do not assume protections always apply or always bypass.
   A ruleset exception needs its own explicit maintainer authorization.
   Save the complete configuration privately, identify the exact rule,
   choose the narrowest available exception and limit it to the import.
   Verify restoration immediately, including enforcement, checks, reviews
   and bypass actors; if restoration fails, stop and alert the maintainer.
   A release request does not authorize disabling repository-wide protection.
4. **Import and verify.** Compare the imported patch to the validated
   private content, record the actual public SHA, and inspect its CI and
   mutation verdict. A changed patch needs validation again. Resolve a
   survivor before preparing the release, covering all affected paths.
   Put the repository's GHSA URL in the `security` fragment; `prepare.sh`
   accepts it without inventing a public PR number, even while it is a draft.
5. **Release, verify, then disclose.** Follow the normal candidate guard,
   merge, tag and publication checks. Only after the fixed assets and
   installation paths are verified publish the advisory with the version
   actually shipped. Verify its `published_at`, patched versions and link.
   Publish or update the threat model and other previously embargoed details
   afterwards. Public documentation that already exists can be maintained,
   but new vulnerability details remain private until disclosure.

Close the record with the GHSA, private and imported patch SHAs, validation
commands/logs/verdicts, actual patched version, publication timestamp and
any exception/restoration evidence. List each follow-up PR, owner and state;
"open" is not "done". Keep one current next action so another session can
resume without following obsolete instructions.

Reference: GitHub's [temporary private fork documentation](https://docs.github.com/en/code-security/tutorials/fix-reported-vulnerabilities/collaborate-in-a-fork).
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: 467kUpdated Claude CodeCodexVerifiedInstall
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: 196kUpdated Claude CodeCodexVerifiedInstall
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: 196kUpdated Claude CodeCodexUnverifiedInstall
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: 196kUpdated Claude CodeCodexVerifiedInstall