quick
Skill2 starsChanged 24 days agoMITView on GitHub
The quick lane of AGENTS.md §12 — a small, settled change from issue to PR without stops, every check kept
- Commits and pushes
18 other projects use a version of this.
Adopted by
Most recently taken up by
- AlleyBo55/no-overthinktheir own version, first seen Nov 8, 2026
- edwinidrus/dragonhoardtheir own version, first seen Nov 8, 2026
- notque/vexjoy-agenttheir own version, first seen Nov 7, 2026
- buildomator/buildomatortheir own version, first seen Nov 4, 2026
- joetawil7/first-passtheir own version, first seen Nov 1, 2026
- 0kim0bos/yamtheir own version, first seen Oct 29, 2026
Used with
Projects with it also use plant-seed, discuss and 4 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.
notqueSkillplant-seed
Capture forward-looking idea as a seed for future feature design.
Also in 3 projects that use this
markmdevSkilldiscuss
Discuss a problem with other AI agents before implementing. Spawns one or more autonomous claude -p conversations with different perspectives (architect, critic, domain expert), runs multi-round discussions, and synthesizes findings. Use when facing design decisions, architecture tradeoffs, knowledge organization questions, or any problem that benefits from multiple viewpoints before committing to an approach. Also use proactively when you're uncertain about a direction.
Also in 4 projects that use this
VercelSkillnext
Next.js renderer for json-render that turns JSON specs into full Next.js applications with routes, layouts, SSR, and metadata. Use when working with @json-render/next, building Next.js apps from JSON specs, or creating AI-generated multi-page applications.
Also in 3 projects that use this
NVIDIASkillresearch
Before designing a feature, research how comparable libraries solve the problem, check best practices, known pitfalls, and upstream changes. Use when starting a new feature, investigating an unfamiliar area, or the user asks to research something.
Also in 4 projects that use this
AWSSkillship
Finalize dev work on an issue branch — update docs, push, and open a PR.
Also in 3 projects that use this
Hugging FaceSkillimplement
Make tests pass. Invoke after /write-tests produces failing tests.
Also in 3 projects that use this
Other versions18 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 18
- notque/claude-code-toolkit★ 448same name
- notque/vexjoy-agent★ 448same name
- govctl-org/govctl★ 276same name
- joetawil7/first-pass★ 128same name
- buildomator/buildomator★ 102same name
- rizukirr/vibekit★ 20same name
- jamillazarev/multica-ops★ 11same name
- codebycorey/delta-spec★ 4same name
- Yafan-Yang/ai-trading★ 4same name
- AlleyBo55/no-overthink★ 3same name
8 more not listed
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
The file
We split this skill into the rules it gives an agent
1 rule. 0 are common, 1 is distinctive, and 0 are very specific to this project.
0
Common
Most of the 15 edited versions we read keep these.
1
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
- Distinctive: line 19
Run the commands from the Bash tool: PowerShell 5.1 mangles the quotes in the arguments of external programs.
“Common” is measured: kept by at least 6 in 10 of the 15 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.
Files
Quick
The quick lane of AGENTS.md §12: the cycle of a small, settled change, from the issue to the PR. Only the user
launches this skill, and launching it is the go-ahead for the issue, the branch, the commit, the push and the PR
of this change: nothing more, no merge.
The lane removes the stops, not the checks. deno task verify, /review, Codex and CI run as in the normal
cycle, and you stop only when one of them has something to say. The prompts of .claude/settings.json on
gh issue create, git push and gh pr create stay: they are the only questions on the way. The body of the
issue, the name of the branch and the commit message are not confirmed one by one: the user reads them in the PR,
before the merge.
Run the commands from the Bash tool: PowerShell 5.1 mangles the quotes in the arguments of external programs.
Source
The change is $ARGUMENTS, read together with this conversation; if empty, the change agreed in this conversation. If there is none, ask what the change is, and stop.
Boundary
The four conditions of AGENTS.md §12, Quick lane. Check them twice:
- At launch, on the description, before anything is created. Condition 1 takes judgment: the issue records decisions, it does not make them. If writing it means picking between alternatives the description leaves open, that pick is the open decision.
- Before the commit, on the diff:
git status --porcelain --untracked-files=normal, which lists new files whatever the Git configuration of the PC, andgit diff. Every file listed counts, changes already in the working tree at launch included.
A condition that fails is an exit: name it, by its number and in words.
The lane
In this order, each step as the normal cycle of §12 runs it:
- The boundary, on the description.
- The issue, written from the change, following the Title and body and the Publish sections of
.claude/skills/issue/SKILL.md. - The branch of the new issue, as §12, Rules, describes it.
- The change. For a change of behavior,
/tdd, from the Acceptance of the new issue (§9). A change with no behavior for a test to check, such as a fix to a document, is written directly. deno task verify, run by you: the Stop hook runs it only when the turn ends, and the lane does not end its turn before the commit. If it is red, fix it and run it again, once: the one retry the Stop hook allows./review, on the uncommitted changes. A report with no finding, only the status of lint and of the invariants, lets the lane go on.- The boundary, on the diff.
- The commit of every file of the diff: in English, Conventional Commits, with the why in the body (§12, Rules).
/codex-review, as it is. With no P0/P1 finding its record getsnoneunder Outcomes, and the lane goes on: the launch is the go-ahead for the push and the PR that its last line leaves to the user.- The PR, following the Title and body and the Publish sections of
.claude/skills/pr/SKILL.md. CI and the merge are the user's (§12, step 8).
Exits
The lane ends at the first of these, and the work goes on in the normal cycle of §12 from the step named here. Say which step that is, and what already exists: the issue with its URL, the branch, the commit.
- The boundary fails at launch: nothing is created. Name the condition, and for condition 1 the open
decision. Step 1: the plan,
/grill-meor/issue, as the change needs. /tddneeds to ask the user, or so does a change written directly: step 2, from the question.deno task verifystays red after the retry: show its output, or its end when it is long. Step 2./reviewreports a finding: step 3. The user decides on the findings, then on the commit.- The boundary fails on the diff: name the condition and the files. The issue and the branch stay, and the changes stay uncommitted. Step 4: the commit waits for the user's go-ahead.
- Codex reports a P0/P1: step 6, as
/codex-reviewdescribes it. Then the user launches/pr.
A prompt the user denies, or a step that stops by its own rules, ends the lane the same way, at that step.
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
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.

