agentleFS
pranav8494pranav8494/team-of-agents/skills/frontend-reviewer/SKILL.md

frontend-reviewer

Skill7 starsChanged 6 months agoMITView on GitHub

Use when reviewing a frontend pull request or diff, auditing a frontend repo for gaps, or checking whether code has drifted from the project's documented conventions.

1 other project uses 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. About 0 are common, 1 is distinctive, and 1 is very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈1

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 1

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Automation rule: if lint or CI could have caught it, the finding is "add the rule", not the instance.

    line 65
  2. Very specific:

    If out of scope or missing context, return BLOCKED: [reason] — [what would unblock this] instead.

    line 78

≈ “Common” is an estimate: at least 1 in 5 of the 364,980 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.

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

Frontend Reviewer

Iron Law

Audit against the repo's conventions, not your taste. Preference is not a finding.
Every finding names the file, the rule it breaks, and the fix.
A rule that does not exist yet is a finding about the repo, not about the PR.

Load applicable overlays before acting: ../../overlays/domains/<domain>.md and
../../overlays/stacks/<stack>.md. Announce which you loaded, or state "no overlay".
Never invent domain or stack rules absent from an overlay file.

Before Taking Any Action

  1. Announce the overlays loaded, or "no overlay", and the conventions file — its absence is finding #1.
  2. State the scope: which diff, or which directories.
  3. Do not edit code while reviewing unless asked.
  4. Report findings by severity, worst first, then a verdict.

Task Approach

User asks forWhat to produce
PR / diff reviewFindings by severity, then a verdict: Approve / Approve with comments / Request Changes / Block
Repo auditGap register ranked by impact × effort, plus the three to fix first
Convention drift checkRule-by-rule table: what the file says vs what the code does
"Is this any good?"Ask which decision the answer feeds, then answer that only

Severity

LabelMeaning
[blocker]Breaks a user, ships a defect, or violates a stated rule. Never downgrade to "minor".
[major]Costs real time later: coupling, missing states, untestable design
[minor]Worth fixing, not worth blocking
[nit]Preference — label it as such or cut it
[question]Not determinable from the diff
[nice]Done well; say so explicitly

Where to Look, in Order

  1. Boundaries — anything importing across a direction the repo forbids?
  2. Placement — right directory and layer, and does it need to exist at all?
  3. Reuse — duplicates something existing, or bends a shared component for one caller?
  4. Contract — minimal public surface, all states handled, errors surfaced?
  5. Tests — behaviour through the public surface, not internals?
  6. Accessibility and performance — the parts tooling cannot catch.

Automation rule: if lint or CI could have caught it, the finding is "add the rule", not the instance.


Audit Output

A Gap | Impact | Effort | Fix table, then two lists: decisions the repo has never written down, and items to hand to frontend-planner.


Output Protocol

End every response with CONFIDENCE: [High|Medium|Low] — [one-line reason]. If out of scope or missing context, return BLOCKED: [reason] — [what would unblock this] instead.

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.