Vercel Composition Patterns
supabase/supabase
React composition patterns that scale. An agent skill from supabase/supabase.
Prepare, publish, recover, or verify OpenClaw beta, stable, and extended-stable releases, including approved backports.
$ npx skills add openclaw/openclaw --skill release-openclaw-maintainer -a claude-codeProject install by default; add -g for ~/.claude/skills/.
$ gh skill install openclaw/openclaw release-openclaw-maintainer --agent claude-codeProject scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).
$ git clone --depth 1 https://github.com/openclaw/openclaw.git skills-src && mkdir -p .claude/skills && cp -r skills-src/.agents/skills/release-openclaw-maintainer .claude/skills/release-openclaw-maintainer && rm -rf skills-srcUse ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.
Claude Code skills documentation · loads skills from .claude/skills/
Install the "release-openclaw-maintainer" agent skill from https://github.com/openclaw/openclaw/tree/main/.agents/skills/release-openclaw-maintainer into .claude/skills/release-openclaw-maintainer/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "release-openclaw-maintainer", then confirm the skill loads.Claude Code copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$skill-installer install https://github.com/openclaw/openclaw/tree/main/.agents/skills/release-openclaw-maintainerType this inside Codex. $skill-installer <name> installs a curated skill from openai/skills. The installer writes to $CODEX_HOME/skills (default ~/.codex/skills). Restart Codex if the skill does not show up.
$ npx skills add openclaw/openclaw --skill release-openclaw-maintainer -a codexProject install goes to .agents/skills/; add -g for ~/.codex/skills/.
$ gh skill install openclaw/openclaw release-openclaw-maintainer --agent codexProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/openclaw/openclaw.git skills-src && mkdir -p .agents/skills && cp -r skills-src/.agents/skills/release-openclaw-maintainer .agents/skills/release-openclaw-maintainer && rm -rf skills-srcUse ~/.agents/skills/ instead of .agents/skills for a personal install.
Codex skills documentation · loads skills from .agents/skills/
Install the "release-openclaw-maintainer" agent skill from https://github.com/openclaw/openclaw/tree/main/.agents/skills/release-openclaw-maintainer into .agents/skills/release-openclaw-maintainer/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "release-openclaw-maintainer", then confirm the skill loads.Codex copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ npx skills add openclaw/openclaw --skill release-openclaw-maintainer -a cursorProject install goes to .agents/skills/; add -g for ~/.cursor/skills/.
$ gh skill install openclaw/openclaw release-openclaw-maintainer --agent cursorProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/openclaw/openclaw.git skills-src && mkdir -p .cursor/skills && cp -r skills-src/.agents/skills/release-openclaw-maintainer .cursor/skills/release-openclaw-maintainer && rm -rf skills-srcUse ~/.cursor/skills/ instead of .cursor/skills for a personal install.
Cursor skills documentation · loads skills from .cursor/skills/, .agents/skills/, .claude/skills/, .codex/skills/
Install the "release-openclaw-maintainer" agent skill from https://github.com/openclaw/openclaw/tree/main/.agents/skills/release-openclaw-maintainer into .cursor/skills/release-openclaw-maintainer/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "release-openclaw-maintainer", then confirm the skill loads.Cursor copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ gemini skills install https://github.com/openclaw/openclaw.git --path .agents/skills/release-openclaw-maintainer--scope user (default) or --scope workspace; --path is the subfolder of the repo that holds the skill; --consent skips the security confirmation prompt.
$ npx skills add openclaw/openclaw --skill release-openclaw-maintainer -a gemini-cliProject install goes to .agents/skills/; add -g for ~/.gemini/skills/.
$ gh skill install openclaw/openclaw release-openclaw-maintainer --agent gemini-cliProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/openclaw/openclaw.git skills-src && mkdir -p .gemini/skills && cp -r skills-src/.agents/skills/release-openclaw-maintainer .gemini/skills/release-openclaw-maintainer && rm -rf skills-srcUse ~/.gemini/skills/ instead of .gemini/skills for a personal install, then run /skills reload.
Gemini CLI skills documentation · loads skills from .gemini/skills/, .agents/skills/
Install the "release-openclaw-maintainer" agent skill from https://github.com/openclaw/openclaw/tree/main/.agents/skills/release-openclaw-maintainer into .gemini/skills/release-openclaw-maintainer/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "release-openclaw-maintainer", then confirm the skill loads.Gemini CLI copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ gh skill install openclaw/openclaw release-openclaw-maintainerInstalls for Copilot at project scope by default; add --scope user for a personal install. Preview a skill first with gh skill preview. Needs GitHub CLI 2.90.0 or later (public preview).
$ npx skills add openclaw/openclaw --skill release-openclaw-maintainer -a github-copilotProject install goes to .agents/skills/; add -g for ~/.copilot/skills/.
$ git clone --depth 1 https://github.com/openclaw/openclaw.git skills-src && mkdir -p .github/skills && cp -r skills-src/.agents/skills/release-openclaw-maintainer .github/skills/release-openclaw-maintainer && rm -rf skills-srcUse ~/.copilot/skills/ instead of .github/skills for a personal install. Commit .github/skills so cloud agent and code review can use it.
GitHub Copilot skills documentation · loads skills from .github/skills/, .claude/skills/, .agents/skills/
Install the "release-openclaw-maintainer" agent skill from https://github.com/openclaw/openclaw/tree/main/.agents/skills/release-openclaw-maintainer into .github/skills/release-openclaw-maintainer/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "release-openclaw-maintainer", then confirm the skill loads.GitHub Copilot copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ npx skills add openclaw/openclaw --skill release-openclaw-maintainer -a opencodeOpenCode documents no install command of its own. Project install goes to .agents/skills/; add -g for ~/.config/opencode/skills/.
$ gh skill install openclaw/openclaw release-openclaw-maintainer --agent opencodeProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/openclaw/openclaw.git skills-src && mkdir -p .opencode/skills && cp -r skills-src/.agents/skills/release-openclaw-maintainer .opencode/skills/release-openclaw-maintainer && rm -rf skills-srcUse ~/.config/opencode/skills/ instead of .opencode/skills for a personal install.
OpenCode skills documentation · loads skills from .opencode/skills/, .claude/skills/, .agents/skills/
Install the "release-openclaw-maintainer" agent skill from https://github.com/openclaw/openclaw/tree/main/.agents/skills/release-openclaw-maintainer into .opencode/skills/release-openclaw-maintainer/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "release-openclaw-maintainer", then confirm the skill loads.OpenCode copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
release-openclaw-maintainerPrepare, publish, recover, or verify OpenClaw beta, stable, and extended-stable releases, including approved backports.
Release Openclaw Maintainer is an agent skill from openclaw/openclaw. Prepare, publish, recover, or verify OpenClaw beta, stable, and extended-stable releases, including approved backports.
Its SKILL.md is about 2.9k tokens, which your agent loads only when the skill is triggered. The skill folder holds 12 other files, including reference files (for example `references/backport-discovery.md`, `references/extended-stable-backports.md` and `references/extended-stable-publish.md`).
It sits in Development. The repository describes itself as: The AI that really does things. Any OS. Any Platform. The lobster way. 🦞. The licence is MIT.
Read from SKILL.md and the folder at commit 1eb5970. It shows what the files ask for, not the result of running them.
Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.
From allowed-tools in the SKILL.md frontmatter.
No scripts in the folder and no shell commands in SKILL.md.
From the folder's file list and the shell code blocks in SKILL.md.
No URLs in SKILL.md.
From URLs in SKILL.md, links to its own repository left out.
Names no API keys, tokens, secrets or passwords.
From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.
Release Openclaw Maintainer loads about 2.9k tokens when it runs, and up to ~36k if it reads all its reference files. Until then it costs about 37 tokens; SKILL.md has 1,517 words of instructions outside code blocks.
Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.
The automated check found no risky patterns in SKILL.md.
Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.
The full file from openclaw/openclaw at commit 1eb5970, republished under its MIT licence (© openclaw). 1,517 words, ~2,871 tokens.
.claude/skills/release-openclaw-maintainer/SKILL.md (or your agent's skills folder). This skill also uses 11 other files; get the full folder from GitHub.Use for a release operation, not ordinary development or advisory mutation.
Read docs/reference/RELEASING.md for current policy. Load $release-private
when available before resolving private credential locators or host topology;
credential operations use $one-password.
Read only the references needed for the selected phase:
.33+ Gateway publication: extended-stable publication. Use the shared publisher with extended-stable inputs; its non-Latest GitHub Release carries evidence without native-app or ClawHub publication.$release-openclaw-ci for workflow execution and immutable manifests.$release-openclaw-mac for macOS operations.$openclaw-changelog-update, including its separate approved post-release docs-mirror route. Initial release generation keeps its existing format; docs publication does not run automatically during release. Requested announcements: $release-openclaw-announcement for Discord, $release-tweets for X. Announcements never gate publication and require explicit posting authorization.$verify-release. GHSA operations: $openclaw-ghsa-maintainer only with explicit security-workflow authorization.Every selected validation lane must succeed: Windows and macOS Node and other
normal CI jobs, install smoke, survivor
lanes, update-first-hop-compat*, pack/npm qualification, package integrity,
and Linux/Windows/macOS Gateway checks, including Windows packaged
install/upgrade checks in Release Checks. A cancelled run still blocks. Preserve
first failures and classify each one as below before recovery. Stable
publication requires stable/full evidence, soak, and blocking performance.
Beta-profile evidence cannot authorize stable publication. No lane or soak
waiver can bypass these requirements. All nine Gateway install/upgrade
combinations across Linux, Windows, and macOS are required for all-group
qualification. Preserve identity, provenance, complete evidence, and existing
publication approvals.
Every failed test gets an explicit lead decision, real release blocker or
flake, recorded in the handoff with its evidence: the same SHA passing
elsewhere or on rerun, no relation to the release delta, a runner or infra
signature, a history of the same case flaking, or a pre-existing product bug
that is not a regression (for example the 2026.9.6 "Assign to…" bug). A real
blocker is a regression in shipped bytes or behavior, or an update/install/
publish defect; fix it on the release branch. Rerun a flake on
the same Release SHA with at most two recorded reruns by default, and file a
fix-in-parallel issue or PR on main with the evidence. Never re-cut, change
tooling, or start a new FRV for a flake. A flake that remains red blocks publication. Main-only failures and infrastructure
failures (runner outages, GitHub ghost jobs, hosted-runner offload) count as
flakes for the release.
Dependency advisories never delay a release. A newly published advisory is
never a reason to re-cut, change tooling, or rerun validation. Record it in the
release evidence and handoff, then file or queue the dependency bump on main
as a normal follow-up after publication. Only a known-malware finding stops
publication. Release dependency evidence and release-dispatched CI enforce this.
Main's CI health never gates a release. Validation and publication run from the
release branch plus pinned tooling, so a red main is not a reason to wait,
re-cut, or pause. When the release needs a release-tooling fix on main
(tooling SHAs must be trusted main commits), main failures the fix does not
cause must not hold that landing: prove on a clean main checkout that the
failure already exists there, record it in the PR body, and merge. During an
active release, an admin merge is allowed for a release-tooling PR in exactly
that situation. Red main still gets fixed, in parallel by a separate lane,
never on the release's critical path.
The operating objectives are approximately 20 minutes to seal validation and
publication within an hour, not measured guarantees. Source-only children start
alongside artifact producers; candidate consumers start as soon as the candidate
is ready. Independently sealed green children can be reused for the same exact
target and inputs even when their parent failed, was cancelled, or remains active;
verify their original admitted qualification identity (or historical trusted-main
workflow SHA) and current attempt. The sealed
manifest supplies the SDK evidence digest and npm publication decisions; it
never acknowledges SDK API changes, so supply
plugin_sdk_api_acknowledgement whenever the SDK report contains changes. The
publisher cannot accept waived validation evidence. Explicit
publisher inputs select publication scope; the candidate helper still validates its explicit
SDK acknowledgement when needed.
Explicit approval is required for version changes and irreversible publication. A request to cut, publish, or complete a named release carries through its validated publication and verification; do not ask again unless identity, channel, scope, or material risk changes. Ship authority for ordinary code is not release authority.
An operator's explicit approval to do whatever is needed to prepare a named release is standing authority for the necessary preparation decisions and repairs. Carry it through candidate and tooling fixes, upgrade/migration design, reviewed test or security-inventory alignments, isolated proof, commits, pushes, and validation recovery. Record the decision, its evidence, and the selected support contract; do not ask again merely because an already-approved class of work reaches an implementation or verification step. Continue independent work while resolving a blocker. This authority does not permit hiding defects, lowering a gate to manufacture success, destructive changes to operator state, unrelated work, or publication. A prepare-only request still requires a separate publication instruction before releasing artifacts or a bridge version.
Keep one compact state record using the handoff template: effective goal, version/tag/branch, cut/Code/Release SHAs, C/Q/P identities, active parent run and attempt, successful child artifacts, approved changes, phase and next action. Latest operator steering replaces superseded scope. Completed evidence stays complete until a named change invalidates it.
For regular releases, prepare complete notes before freezing Code SHA when
possible. If those notes are final, Code SHA and Release SHA are the same
commit: one successful fresh full qualification can supply both roles and
their exact publication bytes. Do not create another commit or run solely to
separate the labels. If notes change after qualification, a descendant whose
complete delta includes CHANGELOG/<version>.md (exact beta version or stable
base) and only that entry, its
matching record, and root index may use split-changelog-release-v1
to reuse product proof while qualifying new publication bytes. Any other
source delta, rename, or deletion returns to the Code SHA loop. Historical
root-only receipts retain changelog-only-release-v1.
New qualification freezes Q=C: the entire qualification workflow closure
belongs to the candidate. Keep independently trusted P (admission, verifier,
and publisher) separate. A Q harness repair requires a new C/Q and newly bound
evidence; P-only or infrastructure repairs can preserve the candidate. Missing
qualification contracts need a deliberate backport, never future-main fallback.
Once a candidate is cut, its base is the operator's decision. Never re-cut
(re-base the candidate on newer main) unless Peter explicitly asks for it in
that release. Without asking, cherry-pick already-merged main commits onto
the release branch only to fix a confirmed release blocker: a required lane
failing deterministically on the frozen candidate, or an update/install/
publish-bytes defect. Name each cherry-pick in the handoff record. Not allowed:
opportunistic backports, feature reverts, or a new base taken to "pick up" a
fix that cherry-picks cleanly enough with a small conflict resolution.
Release process improvements made during a release land on both branches.
Workflow, release-script, release-test, RELEASING.md, and release-skill changes
merge to main first, then get cherry-picked (-x) onto release/YYYY.M.PATCH
for the next candidate or patch. A qualification repair needed for this release
must land before freezing a new C/Q; never silently swap the harness for an
already qualified candidate. Where main-only CI infrastructure is missing on the branch,
keep the branch's expression form and port only the logic. Product code on the
release branch stays blocker-only per the rule above.
A release is not done while anything opened for it is still open. Before the
final report, list every PR created during the release (gh pr list --author @me --state open plus any PR bound to the session) and land or explicitly close
each one with a reason; confirm its fix is on main and, when it is release
tooling, on the release branch. Also remove the release's temporary worktrees,
abandoned local cut branches, and stale scripts/pr worktrees.
Published versions and final tags are immutable. Reuse successful exact-source artifacts; do not rebuild or republish as an implicit retry. The active release is the work queue: no opportunistic moving-main fixes or backports. Classify failures, repair their owner, retry the affected surface, then reassess rather than repeating the full release.
Required publication proofs and enforced environment approvals remain required. A passing sibling cannot replace missing required evidence. npm + ClawHub is the priority path. macOS, Windows, Linux, and Android native publication runs in parallel and never gates npm/ClawHub, GitHub release finalization, or main closeout. Selected Windows Node, Linux/Windows/macOS Gateway, and native-app CI failures block release validation. Platform publishers retain their own artifact and updater contracts; report pending platforms and proof gaps accurately.
© openclaw, MIT. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file
SKILL.md and 11 other files (references) in .agents/skills/release-openclaw-maintainer of openclaw/openclaw.
Open the folder on GitHubat commit 1eb5970
Release Openclaw Maintainer next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.
| Skill | Stars | Used in | Tokens | Auto-check | Licence | Repo updated |
|---|---|---|---|---|---|---|
| Release Openclaw Maintainer this skillopenclaw/openclaw | 392k | — | ~2.9k | Automated safety check: Pass | MIT | |
| Vercel Composition Patternssupabase/supabase | 111k | 59 repos | ~726 | Automated safety check: Pass | MIT | |
| Finishing a Development Branchobra/superpowers | 296k | 5 repos | ~1.9k | Automated safety check: Pass | MIT | |
| Typescript Advanced Typesrolling-scopes/rsschool-app | 10k | 25 repos | ~4.2k | Automated safety check: Pass | MPL-2.0 | |
| PR Babysitteropeninterpreter/openinterpreter | 69k | 3 repos | ~4.2k | Automated safety check: Pass | Apache-2.0 | |
| Code Review ChecklistshareAI-lab/learn-claude-code | 78k | 5 repos | ~1.1k | Automated safety check: Pass | MIT |
supabase/supabase
React composition patterns that scale. An agent skill from supabase/supabase.
obra/superpowers
Walks the last step of a branch: confirm tests pass, detect the git environment, ask how to integrate, carry out your choice and clean up the worktree.
rolling-scopes/rsschool-app
Master TypeScript's advanced type system including generics, conditional types, mapped types, template literals, and utility types for building type-safe applications.
openinterpreter/openinterpreter
Watches an open GitHub pull request until it merges, handling review comments, diagnosing CI failures and retrying flaky checks along the way.
shareAI-lab/learn-claude-code
Reviews code against a five-part checklist covering security, correctness, performance, maintainability and testing, and reports findings in a fixed format.
onyx-dot-app/onyx
Iteratively improves a PR (GitHub), MR (GitLab), or shelved changelist (Perforce) until Greptile gives it a 5/5 confidence score with zero unresolved comments.
openclaw/openclaw
Maintain the canonical live OpenClaw main checkout, macOS LaunchAgent-managed Gateway, local macOS app, exact-head main CI, and recurring full release validation.
openclaw/openclaw
Control tmux sessions/panes for interactive CLIs: list, capture output, send keys, paste text, monitor prompts.
openclaw/openclaw
Feishu document read/write workflows. An agent skill from openclaw/openclaw.
openclaw/openclaw
Review, triage, repair, or land OpenClaw issues and pull requests with current-source evidence and the native maintainer workflow.
openclaw/openclaw
A skill your agent uses when controlling web pages with the OpenClaw browser tool, especially multi-step flows, login checks, tab management, or recovery from stale refs/timeouts.
openclaw/openclaw
A skill your agent uses for all ClawSweeper work: OpenClaw issue/PR sweep reports, repair jobs, cloud fix PRs, @clawsweeper maintainer mention commands, trusted ClawSweeper-reviewed…
Categories
Prepare, publish, recover, or verify OpenClaw beta, stable, and extended-stable releases, including approved backports. Release Openclaw Maintainer is an agent skill from openclaw/openclaw. Prepare, publish, recover, or verify OpenClaw beta, stable, and extended-stable releases, including approved backports.
Release Openclaw Maintainer fits situations like: development work in your project.
Run `npx skills add openclaw/openclaw --skill release-openclaw-maintainer -a claude-code`. Or copy the skill folder (.agents/skills/release-openclaw-maintainer in openclaw/openclaw) into .claude/skills/release-openclaw-maintainer in your project. Claude Code loads it when a task matches its description.
Run `npx skills add openclaw/openclaw --skill release-openclaw-maintainer -a codex`. Or copy the skill folder (.agents/skills/release-openclaw-maintainer in openclaw/openclaw) into .agents/skills/release-openclaw-maintainer in your project. Codex loads it when a task matches its description.
Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add openclaw/openclaw --skill release-openclaw-maintainer -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/release-openclaw-maintainer, .gemini/skills/release-openclaw-maintainer, .github/skills/release-openclaw-maintainer and .opencode/skills/release-openclaw-maintainer in your project.
SKILL.md names no scripts, command-line tools or credentials: Release Openclaw Maintainer is instructions for the agent only.
SKILL.md contains no URLs. Any network use would come from the scripts or tools the agent runs. This is read from the text; nothing was executed.
Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.
Release Openclaw Maintainer is published under the MIT licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.
About 2.9k tokens (SKILL.md is roughly 11k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full. Its references folder adds about 33k tokens, read only when the agent opens those files.
Skills that share tags, products or a category with Release Openclaw Maintainer: Vercel Composition Patterns (supabase/supabase, 111k stars), Finishing a Development Branch (obra/superpowers, 296k stars), Typescript Advanced Types (rolling-scopes/rsschool-app, 10k stars) and PR Babysitter (openinterpreter/openinterpreter, 69k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.
openclaw (a GitHub organization) maintains it in openclaw/openclaw, which has 391,610 GitHub stars. The repository holds 93 skills in this directory. The repository was last updated on October 8, 2026.
Source: openclaw/openclaw on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.