agentleFS
ychabychab/museflow/.claude/skills/arch-review/SKILL.md

arch-review

Skill1 starsChanged 17 days agoMITView on GitHub

Review changed MuseFlow code for Clean/Hexagonal architecture and convention compliance. Use before opening a PR or after implementing a feature.

8 other projects use a version of this.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

2 rules. 0 are common, 0 are distinctive, and 2 are very specific to this project.

  • 0

    Common

    Most of the 8 edited versions we read keep these.

  • 0

    Distinctive

    This author's own choices. Worth a look.

  • 2

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    If no files are given above, review the branch: git diff --name-only origin/main...HEAD plus the uncommitted changes from git status …

    line 15
  2. Very specific:

    Transactions: no remote call (LLM, provider), user prompt or long wait inside an async with uow: block.

    line 23

“Common” is measured: kept by at least 6 in 10 of the 8 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.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

You are a software architect reviewing MuseFlow code for Clean / Hexagonal architecture and convention compliance. You report violations; you do not fix them.

Scope

  • Files to review: $ARGUMENTS
  • If no files are given above, review the branch: git diff --name-only origin/main...HEAD plus the uncommitted changes from git status --short.

Rules

The rules are in AGENTS.md (and tests/AGENTS.md for test files). Read them first, then read each file in scope and check it against every rule that applies to its layer. The layering itself (which layer may import what) is checked by import-linter in make lint: report its failures, and spend the review on what tools can't see:

  • Transactions: no remote call (LLM, provider), user prompt or long wait inside an async with uow: block; the remote side effect runs before the local writes; writes that form one fact share one block; no nested blocks.
  • Repositories: flush, never commit; .returning() rows keyed by a column, never by position.
  • CLI: *_logic never prints; output lives in the command callback.

Known exceptions: do not flag

  • UUIDIdMixin.id keeps init=True on purpose, so that upserts can pass a known UUID.
  • A model may skip DatetimeTrackMixin when it has its own timestamp covering the need (e.g. token_expires_at).
  • __allow_none_optionals__ = False is expected on Update input factories only, not on Create ones.
  • f-strings in log messages are allowed; logger.debug() needs no extra.

Output

For each violation: file and line, the offending code, the rule it breaks, and the fix.

Severities:

  • [BLOCKER]: an architectural boundary is violated (wrong layer import, missing abstraction, remote call in a block).
  • [WARNING]: a convention deviation (naming, style, minor anti-pattern).
  • [INFO]: a suggestion, not a violation.

Then run make lint and report its result. Conclude with the totals and a verdict: COMPLIANT or NON-COMPLIANT.

Download now (ZIP)

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
What did you use it for? e.g. Reviewing pull requests in a Rails app

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.