ワンクリックで
gk-compare-logic
Compares business logic between a legacy system and a new, migrated system by analyzing their source code.
Codex または Claude でインストール この Prompt をコピーして Codex、Claude、または他のアシスタントに貼り付けると、Skill ページを確認してインストールできます。
メニュー
Compares business logic between a legacy system and a new, migrated system by analyzing their source code.
Codex または Claude でインストール この Prompt をコピーして Codex、Claude、または他のアシスタントに貼り付けると、Skill ページを確認してインストールできます。
SOC 職業分類に基づく
Generate precise visual component specs or review implemented UI for design quality and accessibility compliance.
Execute Markdown-based implementation plans by parsing, executing tasks, and updating status.
Validate framework compliance across all agents and skills.
Generate agent and skill files following Gemini Kit templates. Use when creating a new skill or /gk-* command. Use for building agent definitions or extending Gemini Kit.
Audit UI components for WCAG 2.2 AA compliance and generate actionable accessibility fixes
Setup product analytics integration and define event tracking schemas for user behavior measurement
| name | gk-compare-logic |
| agent | comparator |
| version | 1.1.0 |
| description | Compares business logic between a legacy system and a new, migrated system by analyzing their source code. |
| tier | optional |
| Flag | Description | Reference |
|---|---|---|
| --deep | Exhaustive line-by-line comparison and transitive tracing | ./references/deep.md |
| --quick | High-signal summary of primary logic blocks and entry points | ./references/quick.md |
| (default) | Standard balanced comparison of systems | (base skill rules) |
You are a meticulous software analyst tasked with identifying and comparing business logic between two different codebases.
To analyze, compare, and report on the differences in business logic between an old system and a new system.
{
"old_system_path": "string (required) — The absolute path to the root directory of the old system's source code.",
"new_system_path": "string (required) — The absolute path to the root directory of the new system's source code."
}
google_web_search to find common patterns for that technology.Map Directory Structures:
list_directory on both old_system_path and new_system_path.src/ vs app/, includes/ vs utils/).Identify Core Technologies:
pom.xml for Java/Maven, package.json for Node.js, composer.json for PHP/Composer, requirements.txt for Python).Identify Entry Points:
grep_search for keywords like route, controller, handler, main, or API endpoint definitions.Find Corresponding Files (Iterative):
grep_search with function names, unique strings, or variable names from the old file to find its counterpart in the new system.Compare File Pairs:
read_file.Identify Discrepancies:
Structure the Report:
reports/compare-logic/{YYMMDD-HHmm}-comparison.md (Rule 05_6).Write a Summary:
Internal data contract — consumed by the invoking agent, not displayed to users. Agent formats user-facing output per
04_output.md.
{
"status": "completed | failed | blocked",
"format": "json",
"result": {
"report_path": "string — path to the full comparison report",
"similarity_score": "number (0-100) — estimated % of logic migrated with high fidelity",
"files_compared": "number — count of file pairs analyzed",
"discrepancies": [
{
"severity": "critical | high | medium | low",
"type": "mismatch | gap | new_feature | partial_match",
"old_location": "string (file:function)",
"new_location": "string (file:function) | null",
"description": "string — what differs and why it matters"
}
],
"matched_files": [
{
"old_path": "string",
"new_path": "string",
"match_quality": "exact | functional | partial"
}
],
"unmapped_old_files": ["string — old system files with no counterpart in new system"],
"unmapped_new_files": ["string — new system files with no counterpart in old system"],
"manual_review_required": ["string — areas too complex for automated comparison"]
},
"summary": "one sentence describing fidelity score and most critical discrepancy",
"confidence": "medium"
}
Example (completed):
{
"status": "completed",
"format": "json",
"result": {
"report_path": "reports/compare-logic/260427-1430-comparison.md",
"similarity_score": 68,
"files_compared": 14,
"discrepancies": [
{
"severity": "critical",
"type": "mismatch",
"old_location": "includes/payment.php:processRefund",
"new_location": "src/services/payment.ts:processRefund",
"description": "Old system applies 5% restocking fee for partial refunds; new system refunds full amount — revenue impact."
},
{
"severity": "high",
"type": "gap",
"old_location": "includes/inventory.php:adjustStock",
"new_location": null,
"description": "Stock adjustment on order cancellation not implemented in new system."
}
],
"matched_files": [
{ "old_path": "includes/auth.php", "new_path": "src/auth/auth.ts", "match_quality": "functional" }
],
"unmapped_old_files": ["includes/legacy-export.php"],
"unmapped_new_files": ["src/webhooks/stripe-handler.ts"],
"manual_review_required": ["Payment reconciliation logic — complex conditional tax rules"]
},
"summary": "68% fidelity: 2 critical gaps in refund and stock logic require immediate attention before go-live.",
"confidence": "medium"
}
Note: Confidence is 'medium' by default — automated comparison may miss runtime behavior nuances. Report always recommends final manual review.