Ready to supercharge your AI agents?

Install any skill in seconds. Free to start, no credit card required.

Get Started Free →

Refactoring Assistant Skill · a5c-ai | DecimalAI

a5c-ai/refactoring-assistant

SkillScore73Maintained

Suggest and guide application of refactoring patterns to improve code quality

codingby a5c-ai

Links this skill to your agents and keeps it current with the author’s updates.

Automatic

Coding pack

1 · One-time — install the CLI (docs ↗)

pip install decimalai

2 · Get the skill — writes .claude/skills/a5c-ai-refactoring-assistant/SKILL.md

decimalai skills pull a5c-ai-refactoring-assistant --out .claude/skills/

No CLI? Get the same file — curl · raw file

curl -fsSL --create-dirs -o .claude/skills/a5c-ai-refactoring-assistant/SKILL.md https://app.decimal.ai/s/a5c-ai-refactoring-assistant/SKILL.md

Raw file: /s/a5c-ai-refactoring-assistant/SKILL.md ↗ · machine-readable summary at /s/a5c-ai-refactoring-assistant.json ↗

In productionGlobal

how this skill behaves on real traffic · last 30 days

No production data for this skill yet — usage numbers appear here once agents start running it.

Anonymized aggregate across every org using this skill — proof from real production use, not a demo.

Does this skill help? · Benchmark: with vs. withoutsign in to run this on your agents

22 test cases from the publisher’s eval.yaml · each run twice on gemini-3.6-flash — with the skill injected, then without · last run 53d ago

Yes — +59% lift · at higher cost+41% tokens0% turns73% of whole cases pass with skill (16/22)

Test case	Without → With	Effect	Δ tokens	Δ turns
case-13	✗→✓	▲ Improved	58%	0%
case-14	✗→✓	▲ Improved	88%	0%
case-07	✗→✓	▲ Improved	29%	0%
case-08	✗→✓	▲ Improved	103%	0%
case-09	✗→✓	▲ Improved	118%	0%

Showing 5 of 22 cases

SkillSafety

✓ 3 of 3 passed

Code scan15 of 15 checks clean

0 critical · 0 warning · 0 info — scanner v4 — 7/31/2026

live_secretremote_code_executionreverse_shelldata_exfiltrationssrf_metadataagent_snoopinganti_refusaldestructive_commandsobfuscated_unicodeunicode_homoglyphopaque_blobsuspicious_urltool_overreachtrigger_abuseprompt_injection_phrasing

IntentSafe

LLM judge · does the skill do what it claims

ContentClean

LLM judge · policy & harmful-content review

Source

Imported

a5c-ai / babysitter

library/specializations/code-migration-modernization/skills/refactoring-assistant/SKILL.md

View on GitHub ↗

License MITImported 7/31/2026Version v1

Skill contents

SKILL.md

Refactoring Assistant Skill

Suggests appropriate refactoring patterns and guides their application to improve code quality, maintainability, and migration readiness.

Purpose

Enable guided refactoring for:

Code quality improvement
Design pattern application
Migration preparation
Technical debt reduction
Maintainability enhancement

Capabilities

1. Extract Method/Class Suggestions

Identify extraction candidates
Suggest method boundaries
Recommend class decomposition
Guide interface extraction

2. Move Method Recommendations

Detect misplaced methods
Suggest target classes
Analyze dependencies
Guide safe relocation

3. Rename Refactoring

Suggest meaningful names
Propagate renames safely
Handle cross-references
Update documentation

4. Inline Refactoring

Identify inline candidates
Remove unnecessary indirection
Simplify call chains
Reduce complexity

5. Pull Up/Push Down Member

Analyze inheritance hierarchies
Suggest member movement
Optimize class hierarchies
Improve polymorphism

6. Design Pattern Application

Recognize pattern opportunities
Guide pattern implementation
Suggest appropriate patterns
Validate pattern application

Tool Integrations

| Tool | Purpose | Integration Method | |------|---------|-------------------| | IDE Refactoring | Automated refactoring | CLI / API | | Sourcery | AI-powered suggestions | CLI | | Sourcegraph Cody | Code intelligence | API | | IntelliJ IDEA | Java refactoring | CLI | | VS Code | Multi-language | Extension API |

Output Schema

json
{
  "analysisId": "string",
  "timestamp": "ISO8601",
  "suggestions": [
    {
      "type": "extract-method|move-method|rename|inline|design-pattern",
      "priority": "high|medium|low",
      "target": {
        "file": "string",
        "element": "string",
        "line": "number"
      },
      "description": "string",
      "rationale": "string",
      "steps": ["string"],
      "estimatedImpact": {
        "complexity": "string",
        "maintainability": "string",
        "testability": "string"
      },
      "risks": ["string"]
    }
  ],
  "patterns": {
    "applicable": ["string"],
    "opportunities": []
  }
}

Integration with Migration Processes

code-refactoring: Primary refactoring guidance
technical-debt-remediation: Debt reduction strategies

Related Skills

code-smell-detector: Identifies refactoring needs
codemod-executor: Automates refactorings

Related Agents

code-transformation-executor: Executes suggested refactorings

Illustrative Want to see what a5c-ai/refactoring-assistant would put in front of your model? Preview the rendered body and check its trigger phrases — no LLM call, no signup.

Latest benchmark · publisher runVerified

8/2/2026, 20:01 UTC · 22 cases

Lift

+59%

73% vs 14%

Agent turns

0%

1.0 vs 1.0

Tokens

+41%

2,615.23 vs 1,860.82

fail → pass · 13pass → pass · 3fail → fail · 6

Per-case results (22)

Case	Status	Duration (ms)			Turns			Tokens			Tool calls
Case	Status	Without	With	Δ	Without	With	Δ	Without	With	Δ	Without	With	Δ
case-13 Analyze `ShippingService.go` line 60 for deep nesting refactoring. Ensure the JSON response details estimated impacts.	fail→pass	9,231	9,499	+3%	1	1	0%	1,612	2,553	+58%	0	0	—
case-01 We have a legacy Python module `PaymentProcessor.py` where `process_transaction` around line 120 has grown into a 300-line monolith. Please analyze this code and provide a complete refactoring report. I need the response to include analysis metadata, prioritized refactoring suggestions with target files, elements, and line numbers, clear rationales, step-by-step action plans, risk assessments, and impacts on code complexity and testability, alongside any applicable design pattern opportunities.	fail→fail	26,027	18,801	-28%	1	1	0%	4,922	3,745	-24%	0	0	—
case-02 Our Java backend class `OrderService.java` (around line 85) has several misplaced methods and excessive indirection that make unit testing difficult before our upcoming framework upgrade. Analyze the file and give me a detailed refactoring plan outlining high-to-low priority changes, exact targets, steps to execute, potential migration risks, estimated effects on maintainability, and applicable architectural design patterns.	fail→fail	22,214	18,652	-16%	1	1	0%	3,586	4,257	+19%	0	0	—
case-03 I need to refactor `UserAnalytics.ts` around line 50 to simplify call chains and extract helper methods. Please evaluate the code and output a structured analysis that lists target elements, line positions, step-by-step guidance, rationale behind each recommendation, risk factors, impact estimates across testability and complexity, and pattern application opportunities.	fail→fail	16,306	13,959	-14%	1	1	0%	3,029	3,312	+9%	0	0	—
case-14 Evaluate `AuthTokenValidator.js` line 90 for critical security-impacting monolithic logic versus minor variable renaming. Provide JSON analysis.	fail→pass	9,155	17,010	+86%	1	1	0%	1,028	1,937	+88%	0	0	—
case-04 Please run an automated JSCodeShift transformer script on src/components/Button.tsx to convert all class components to functional components and overwrite the file on disk.	fail→fail	5,801	9,973	+72%	1	1	0%	1,021	2,547	+149%	0	0	—
case-05 Run static analysis across my repository to generate a raw count of cyclomatic complexity violations and code smell metrics without generating refactoring plans or suggestions.	fail→fail	7,090	7,383	+4%	1	1	0%	1,079	1,924	+78%	0	0	—
case-06 Write a comprehensive Jest unit test file with mock implementations for UserRegistrationService.ts to achieve 100% code coverage.	fail→fail	15,566	15,830	+2%	1	1	0%	3,794	4,286	+13%	0	0	—
case-07 In `InvoiceCalculator.cs`, `CalculateTotal` at line 45 has 150 lines performing tax calculation, discount evaluation, and PDF formatting. Provide a structured refactoring evaluation in JSON format.	fail→pass	11,818	10,851	-8%	1	1	0%	2,215	2,860	+29%	0	0	—
case-08 In `CustomerController.py`, method `format_billing_address` at line 78 directly manipulates internal attributes of `Address` class. Suggest relocation in JSON format.	fail→pass	5,100	6,770	+33%	1	1	0%	952	1,931	+103%	0	0	—
case-09 `DataHandler.java` uses single-letter variables `d`, `tmp`, `fn` across line 112. Provide structured JSON refactoring guidance to rename these.	fail→pass	6,558	13,032	+99%	1	1	0%	1,314	2,864	+118%	0	0	—
case-10 `OrderPipeline.ts` line 34 contains `validateAndRouteOrder` which only calls `routeOrder` with unchanged parameters. Provide a structured analysis in JSON format to simplify this indirection.	fail→pass	8,455	7,207	-15%	1	1	0%	1,613	2,110	+31%	0	0	—
case-11 `NotificationSender.kt` line 22 uses a 10-branch switch statement based on notification channel types. Suggest replacing conditional logic with a design pattern in JSON format.	fail→pass	7,671	8,385	+9%	1	1	0%	1,549	2,263	+46%	0	0	—
case-12 `Dog.java` and `Cat.java` both declare identical `eat()` methods at line 15. Provide a JSON refactoring plan to hoist this into `Animal.java`.	fail→pass	5,942	6,350	+7%	1	1	0%	1,249	1,897	+52%	0	0	—
case-15 Evaluate `ReportGenerator.rb` line 105 for extracting formatting logic. Target file is `ReportGenerator.rb` and element is `generate_summary`. Provide JSON response.	pass→pass	5,326	7,602	+43%	1	1	0%	778	2,075	+167%	0	0	—
case-16 Evaluate `PaymentGatewayFactory.py` line 40 for creational design pattern opportunities. Output JSON analysis.	fail→pass	9,851	12,587	+28%	1	1	0%	1,692	3,110	+84%	0	0	—
case-17 Analyze `CacheManager.cpp` line 140 where lock acquisition logic is scattered. Output JSON refactoring guidance with action steps and risk assessment.	fail→pass	8,333	7,746	-7%	1	1	0%	1,544	2,050	+33%	0	0	—
case-18 Prepare `LegacyDatabaseAdapter.cs` line 200 for migration to EF Core by breaking down direct SQL building logic. Provide structured JSON.	fail→pass	9,112	9,856	+8%	1	1	0%	1,607	2,489	+55%	0	0	—
case-19 `LegacyUtils.js` line 15 contains multiple single-line pass-through wrappers around standard Math functions. Provide JSON refactoring analysis.	fail→pass	7,686	7,348	-4%	1	1	0%	1,469	2,000	+36%	0	0	—
case-20 `UserProfileManager.ts` line 10 handles authentication, image uploading, and DB persistence (800 lines). Provide structured JSON refactoring guidance.	pass→pass	9,769	8,702	-11%	1	1	0%	1,932	2,250	+16%	0	0	—
case-21 `Vehicle.java` line 25 defines `deployAirbag()` which is only applicable to `Car` subclass and irrelevant to `Bicycle`. Provide JSON refactoring guidance.	pass→pass	8,126	9,460	+16%	1	1	0%	1,664	2,362	+42%	0	0	—
case-22 Provide a refactoring evaluation for `DataPipeline.py` line 75 in JSON format.	fail→pass	7,084	10,620	+50%	1	1	0%	1,289	2,713	+110%	0	0	—

These cases were also screened by the skill’s intent / safety pass.See the Trust review ↗Reproduce this benchmark in Colab ↗

How this skill was measured

DecimalAI ran this skill against gemini-3.6-flash twice over the same eval suite — once with the skill loaded and once without — and compared the two runs case by case. 22 cases were attempted. The headline lift of +59 percentage points is the difference between those two pass rates over the 22 comparable cases. 1 case got worse with the skill loaded, and it is included in that figure.

Model measured: gemini-3.6-flash
Graded by: gemini-3.6-flash
Grading method: judged
Verification: verified
Pass rate without the skill: 13.6%
Pass rate with the skill: 72.7%
Cases attempted: 22
Cases counted toward lift: 22
Cases that regressed: 1
Skill version measured: v1
Run completed: 8/2/2026, 20:01 UTC

Measured example — same prompt, with vs. without the skill · from the benchmark run · case case-13

Prompt

Analyze `ShippingService.go` line 60 for deep nesting refactoring. Ensure the JSON response details estimated impacts.

Without the skill loaded, the model failed this case. With it loaded, the same prompt on the same model passed. This is one improved case from the latest verified run; every case, including any that regressed, is in the table above.

Show both answers

SkillScore blends 4 signals1 of 4 complete

BenchmarkPass rate WITH skill on a verified run73

Live evalNeeds ≥10 evaluated traces in 30 days

AI ratingNeeds ≥10 rated traces

AdoptionNeeds ≥10 serves across ≥2 orgs in 30 days

Efficiency · separate axis (at equal correctness)+0% turns+53% tokens

Shown for transparency; not credited unless a verified run shows no correctness regression.

Backed by 1 of 4 signals — each added signal makes the score more robust.

Verified on gemini-3.6-flash · re-tested 53 days ago

Other measured skills in the registry, with their headline benchmark lift.