All skills
lexler avatar

/refactoring-team

@628d2e9
by Lada Kesselerlexler/skill-factory238 stars
61

Iterative code refactoring through progressive lenses via a worker-reviewer agent team.

Use this Skill: https://skilld.dev/gh/lexler/skill-factory/refactoring-team

This session only. Nothing lands on disk.

referencesreviewer-guides04-abstraction-consistency.reviewer.md

≈491 tokens on demand. Your agent reads this file only when SKILL.md points to it.

Lens: Abstraction Consistency (Reviewer Guide)

When to Apply

After method length. Functions are now smaller, making level mixing visible within functions and design maturity differences visible across modules.

What to Look For in Diffs

  • Inline code extracted to match surrounding abstraction level
  • Entry points becoming pure orchestration
  • Sibling functions unified to follow the same structural pattern
  • Modules solving the same kind of problem brought to consistent design maturity

First Pass

Worker should find within-function mixing — inline detail surrounded by named operations, the squint test failure. Worker should also identify cross-file inconsistency: sibling functions or modules solving the same kind of problem with different design approaches.

Check for false positives: domain code and glue code legitimately differ in shape. Don't force consistency between modules solving genuinely different kinds of problems. Push back if worker standardized too aggressively.

Second Pass

If within-function levels still feel mixed, look at entry points: are they pure orchestration or do they still do detailed work? Find lines where the abstraction level visibly drops — arg parsing, path manipulation, or I/O details sitting among high-level calls.

If cross-file consistency was missed, compare functions that serve the same role: do they follow the same structural pattern? Compare modules doing the same kind of work: does one look designed while the other looks scripted? Check test files for the same interface exercised via different mechanisms.

Check for wrong fixes: if worker extracted one-liners into functions that just restate their body, that's shallow extraction — adding names without hiding complexity. The test: does the function name tell you something its body doesn't?

When Done

Move on when functions read at one level and sibling modules show the same design shape for the same kind of problem.

Source: SKILL.md on GitHub

No alerts6d3 checks · Risk SAFE
  • Gen Agent Trust Hub6d

    The skill facilitates an automated, iterative refactoring workflow using a worker-reviewer agent team. It processes local source code and executes user-provided test commands to verify changes. The skill operates using local scripts and standard development tools without any detected malicious behaviors or obfuscation.

  • Socket6d

    No alerts

  • Snyk6d

    Risk: LOW · No issues

Signed by skilld at 628d2e9. This ties the file your Agent reads to that commit on GitHub. It does not review the instructions.

Last checked against GitHub last month.

Steadyupdated 7 months ago
disable-model-invocation
true
argument-hint
[target-path]
Other metadata
hooks
{
  "TeammateIdle": [
    {
      "hooks": [
        {
          "type": "command",
          "command": "${CLAUDE_SKILL_DIR}/references/guard-idle-worker.sh"
        }
      ]
    }
  ]
}

README badge

README badge for lexler/skill-factory/refactoring-team