arb
Skill2 starsChanged 23 days agoApache-2.0View on GitHub
Reviews architecture proposals, ADRs, design documents, or architecture changes as an Architecture Review Board (ARB). Use when the user invokes $arb, or asks for ARB review, architecture board review, or architecture decision review. Criteria pending; no invented policy or pass/fail verdict.
In this repository13 other files it gives its agents.
- CLAUDE.mdCLAUDE.md
- bro-whatSkill
- debloatSkill
- fitSkill
- security-and-hardeningSkill
- source-briefSkill
- word-ahaSkill
- arb-cnSkill
- debloat-cnSkill
- fit-cnSkill
- principle-thin-wrapper-cnSkill
- source-brief-cnSkill
- word-aha-cnSkill
The file
We split this skill into the rules it gives an agent
2 rules. About 0 are common, 2 are distinctive, and 0 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈2
Distinctive · estimate
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 40
If the subject or evidence boundary is missing and changes the review path, ask one blocking question.
- Distinctive: line 41
Stop after stating the subject, the evidence in hand, and that criteria are pending.
≈ “Common” is an estimate: at least 1 in 5 of the 372,429 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Architecture Review Board
Review a named architecture proposal, ADR, design document, or architecture change as an Architecture Review Board.
Activate only when the user names $arb or this skill.
Agent discipline
Path-changing ambiguity only; no re-asks on stated or settled facts; proceed when context suffices; one blocking question when user judgment is required. Mandatory domain clarifiers (scope, time, IDs) still apply—never as a questionnaire.
Status
Board criteria, decision grades, and the verdict template are not authored yet. State the subject and the evidence in hand, then stop. Inventing ARB policy, scoring, or a substitute review framework is out of scope.
When invoked:
- Identify the review subject from the request and repository evidence.
- If the subject or evidence boundary is missing and changes the review path, ask one blocking question.
- Stop after stating the subject, the evidence in hand, and that criteria are pending.
Completion
Done when the subject is identified, available evidence is listed, and pending criteria are reported. A pass/fail verdict waits until criteria land in this skill.
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.

