Ready to supercharge your AI agents?

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

Get Started Free →

Codemod Executor Skill · a5c-ai | DecimalAI

a5c-ai/codemod-executor

SkillScore42Maintained

Execute automated AST-based code transformations for large-scale refactoring and migration

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-codemod-executor/SKILL.md

decimalai skills pull a5c-ai-codemod-executor --out .claude/skills/

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

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

Raw file: /s/a5c-ai-codemod-executor/SKILL.md ↗ · machine-readable summary at /s/a5c-ai-codemod-executor.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

24 test cases from the publisher’s eval.yaml · each run twice on gemini-3.6-flash — with the skill injected, then without · last run 60d ago

Yes — +42% lift+8% tokens0% turns42% of whole cases pass with skill (10/24)

Test case	Without → With	Effect	Δ tokens	Δ turns
case-01	✗→✓	▲ Improved	-10%	0%
case-02	✗→✓	▲ Improved	19%	0%
case-03	✗→✓	▲ Improved	29%	0%
case-07	✗→✓	▲ Improved	-3%	0%
case-11	✗→✓	▲ Improved	-11%	0%

Showing 5 of 24 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/codemod-executor/SKILL.md

View on GitHub ↗

License MITImported 7/31/2026Version v1

Skill contents

SKILL.md

Codemod Executor Skill

Executes automated code transformations using AST-based tools for large-scale refactoring, framework migrations, and language version upgrades.

Purpose

Enable automated code transformations for:

Framework migration (React, Angular, Vue)
Language version upgrades
API deprecation handling
Pattern standardization
Large-scale refactoring

Capabilities

1. AST-Based Transformations

Parse source code to AST
Apply transformation rules
Generate modified source
Preserve code style

2. Pattern Matching and Replacement

Define source patterns
Specify target patterns
Handle edge cases
Support wildcards and captures

3. Multi-File Transformations

Apply across entire codebase
Handle import updates
Coordinate cross-file changes
Manage side effects

4. Dry-Run Preview

Preview changes before applying
Generate diff reports
Estimate impact scope
Identify risky transformations

5. Rollback Support

Create transformation checkpoints
Enable selective rollback
Track applied codemods
Maintain change history

6. Custom Codemod Creation

Define new transformation rules
Test codemods in isolation
Document transformation logic
Share across teams

Tool Integrations

| Tool | Language | Integration Method | |------|----------|-------------------| | jscodeshift | JavaScript/TypeScript | CLI | | ts-morph | TypeScript | Library | | Rector | PHP | CLI | | Scalafix | Scala | CLI | | OpenRewrite | Java | CLI / Maven | | Bowler | Python | CLI | | ast-grep | Multi-language | CLI / MCP | | gofmt -r | Go | CLI |

Output Schema

json
{
  "executionId": "string",
  "timestamp": "ISO8601",
  "codemod": {
    "name": "string",
    "version": "string",
    "description": "string"
  },
  "results": {
    "filesProcessed": "number",
    "filesModified": "number",
    "filesSkipped": "number",
    "transformations": "number"
  },
  "changes": [
    {
      "file": "string",
      "transformations": [
        {
          "type": "string",
          "line": "number",
          "before": "string",
          "after": "string"
        }
      ]
    }
  ],
  "errors": [],
  "warnings": []
}

Integration with Migration Processes

code-refactoring: Large-scale refactoring
framework-upgrade: Framework migrations
language-version-migration: Syntax upgrades
code-translation: Language conversion assist

Related Skills

refactoring-assistant: Suggests transformations
static-code-analyzer: Pre-transformation analysis

Related Agents

code-transformation-executor: Orchestrates codemods
framework-upgrade-specialist: Framework-specific codemods

Illustrative Want to see what a5c-ai/codemod-executor 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

8/1/2026, 01:34 UTC · 24 cases

Lift

+42%

42% vs 0%

Agent turns

0%

1.0 vs 1.0

Tokens

+8%

1,844.88 vs 1,705.75

fail → pass · 10fail → fail · 14

Per-case results (24)

Case	Status	Duration (ms)			Turns			Tokens			Tool calls
Case	Status	Without	With	Δ	Without	With	Δ	Without	With	Δ	Without	With	Δ
case-01 We need to clean up deprecated lifecycle methods across our codebase in `src/components/`. Please run a jscodeshift AST transformation to update `componentWillReceiveProps` calls. Make sure your output provides an execution identifier, timestamp, codemod metadata, overall results stats (processed, modified, skipped counts), detailed per-file diffs including line numbers with before and after snippets, and arrays for warnings or errors.	fail→pass	10,900	6,848	-37%	1	1	0%	2,422	2,190	-10%	0	0	—
case-02 Please execute an automated rewrite rule on our backend Java repository to migrate old JUnit 4 assertions to JUnit 5 Jupiter. Output the results in a structured JSON schema that includes execution ID, ISO timestamp, details about the codemod used, aggregate file metrics, file-by-file transformation lists with before/after blocks and line references, along with lists for errors and warnings.	fail→pass	11,210	10,227	-9%	1	1	0%	2,565	3,057	+19%	0	0	—
case-03 Run an AST-based automated code transformation to update obsolete logger function names from `logger.warn()` to `logger.warning()` across all Python files. I need the response formatted to show the execution ID, timestamp, codemod info, processing statistics (files processed, modified, skipped, total transformations), explicit change items showing line numbers and old vs new code for each file, and error/warning tracking.	fail→pass	9,838	8,572	-13%	1	1	0%	2,056	2,651	+29%	0	0	—
case-04 We need an automated tool for PHP code upgrades and framework migrations. Which CLI tool is standard for automated PHP AST refactoring?	fail→fail	7,449	3,892	-48%	1	1	0%	1,393	1,434	+3%	0	0	—
case-05 Which CLI tool should we run for automated Scala code refactoring and lint-fixing transformations across our codebase?	fail→fail	10,834	4,116	-62%	1	1	0%	1,986	1,401	-29%	0	0	—
case-06 We want to perform simple rewrite pattern substitutions across our Go source files using native tooling. Which command-line option is used for Go pattern-based rewriting?	fail→fail	3,461	2,852	-18%	1	1	0%	560	1,200	+114%	0	0	—
case-07 Which command-line refactoring tool specialized in Python AST transformations should be used to automate code modernizations?	fail→pass	12,989	5,786	-55%	1	1	0%	1,805	1,748	-3%	0	0	—
case-08 We have microservices in multiple programming languages and need a tool with CLI or MCP integration for structural AST search and replacement. What tool fits this requirement?	fail→fail	13,493	8,728	-35%	1	1	0%	1,874	2,089	+11%	0	0	—
case-09 For automated Java migrations across Maven builds, which transformation tool operates via CLI or Maven plugin?	fail→fail	5,686	2,163	-62%	1	1	0%	1,113	1,082	-3%	0	0	—
case-10 We want to write programmatic TypeScript AST transformation scripts using a dedicated library rather than a CLI framework. Which library fits this integration pattern?	fail→fail	9,844	6,307	-36%	1	1	0%	1,960	1,827	-7%	0	0	—
case-11 Format the execution summary JSON for an AST transformation run on 50 JavaScript files. Provide realistic dummy values for the summary statistics field.	fail→pass	8,084	4,068	-50%	1	1	0%	1,768	1,565	-11%	0	0	—
case-12 Generate a sample changes array entry for a codemod that converted legacy var statements to const on line 14 of src/index.js.	fail→pass	4,463	2,849	-36%	1	1	0%	952	1,137	+19%	0	0	—
case-13 Construct the codemod metadata JSON object block for an automated AST transformation named react-v18-upgrade version 1.2.0 that converts legacy root render syntax.	fail→fail	7,703	5,303	-31%	1	1	0%	1,334	1,694	+27%	0	0	—
case-14 Before executing a massive React version upgrade across 300 files, we want to inspect all changes without altering any source files on disk. What operational mode should be executed?	fail→fail	5,857	4,915	-16%	1	1	0%	915	1,435	+57%	0	0	—
case-15 An automated codemod made unexpected edits to several files during a migration. How should the execution workflow handle reverting changes safely?	fail→pass	12,227	10,435	-15%	1	1	0%	2,059	2,643	+28%	0	0	—
case-16 Our team has internal API deprecations not covered by open-source migration tools. What process should be followed to define and deploy new proprietary AST transformation rules?	fail→fail	18,961	17,152	-10%	1	1	0%	2,588	3,173	+23%	0	0	—
case-17 When moving a shared utility function from src/utils/old.ts to src/helpers/new.ts, how does multi-file codemod transformation handle code references?	fail→fail	15,428	17,475	+13%	1	1	0%	2,548	3,257	+28%	0	0	—
case-18 Which related skill should be used prior to running codemods when we only want suggestions for potential code transformations rather than execution?	fail→pass	12,446	2,175	-83%	1	1	0%	1,581	1,024	-35%	0	0	—
case-19 Which related skill should be invoked before code transformation to perform pre-transformation analysis?	fail→fail	9,226	1,480	-84%	1	1	0%	1,325	890	-33%	0	0	—
case-20 When orchestrating large-scale codemod runs across repositories, which dedicated agent handles orchestrating the codemods?	fail→pass	9,456	3,646	-61%	1	1	0%	1,453	909	-37%	0	0	—
case-21 Which specialized agent is designed for orchestrating framework-specific codemods such as major React or Angular upgrades?	fail→pass	13,801	2,701	-80%	1	1	0%	1,638	1,021	-38%	0	0	—
case-22 Analyze src/auth.ts to detect potential security vulnerabilities, lint issues, and dead code without modifying any files or running codemods. Provide a vulnerability report.	fail→fail	8,328	8,707	+5%	1	1	0%	903	1,509	+67%	0	0	—
case-23 Review this function and suggest cleaner architectural patterns or refactoring ideas without executing AST transforms or applying file edits: ```javascript function handle(x) { if(x) { return x.y; } else { return null; } } ```	fail→fail	8,216	8,664	+5%	1	1	0%	1,518	2,259	+49%	0	0	—
case-24 Write a GitHub Actions YAML workflow that runs npm test and deploys our Docker container to AWS ECS on pushes to main branch.	fail→fail	11,220	12,272	+9%	1	1	0%	2,622	3,082	+18%	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. 24 cases were attempted. The headline lift of +42 percentage points is the difference between those two pass rates over the 24 comparable cases.

Model measured: gemini-3.6-flash
Graded by: gemini-3.6-flash
Grading method: judged
Verification: verified
Pass rate without the skill: 0%
Pass rate with the skill: 41.7%
Cases attempted: 24
Cases counted toward lift: 24
Cases that regressed: 0
Skill version measured: v1
Run completed: 8/1/2026, 01:34 UTC

Measured example — same prompt, with vs. without the skill · from the benchmark run · case case-01

Prompt

We need to clean up deprecated lifecycle methods across our codebase in `src/components/`. Please run a jscodeshift AST transformation to update `componentWillReceiveProps` calls. Make sure your output provides an execution identifier, timestamp, codemod metadata, overall results stats (processed, modified, skipped counts), detailed per-file diffs including line numbers with before and after snippets, and arrays for warnings or errors.

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 run42

Live evalNeeds ≥10 evaluated traces in 30 days

AI ratingNeeds ≥10 rated traces

AdoptionNeeds ≥10 serves across ≥2 orgs in 30 days

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

Verified on gemini-3.6-flash · re-tested 59 days ago

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