mirror of
https://fastgit.cc/github.com/Yeachan-Heo/oh-my-claudecode
synced 2026-04-30 13:51:15 +08:00
3a833c3395
Extend the existing harsh-critic benchmark framework with reusable benchmarks for code-reviewer, debugger, and executor agents. Enables measurable prompt tuning by comparing old (pre-consolidation) vs new (merged) prompts with ground-truth scoring. New infrastructure: - benchmarks/shared/ — generalized scoring types, parser, reporter, runner - benchmarks/code-reviewer/ — 3 fixtures (SQL injection, clean code, payment edge cases) - benchmarks/debugger/ — 3 fixtures (React undefined, Redis intermittent, TS build errors) - benchmarks/executor/ — 3 fixtures (trivial, scoped, complex tasks) - benchmarks/run-all.ts — top-level orchestrator with --save-baseline and --compare modes - npm scripts: bench:prompts, bench:prompts:save, bench:prompts:compare Each benchmark includes archived pre-consolidation prompts for reproducible comparison even after old agent files are deleted. Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>
58 lines
2.6 KiB
JSON
58 lines
2.6 KiB
JSON
{
|
|
"fixtureId": "task-input-validation",
|
|
"fixturePath": "fixtures/tasks/task-input-validation.md",
|
|
"domain": "task",
|
|
"expectedVerdict": "scoped",
|
|
"isCleanBaseline": false,
|
|
"findings": [
|
|
{
|
|
"id": "IMPL-IV-1",
|
|
"severity": "CRITICAL",
|
|
"category": "finding",
|
|
"summary": "Must validate name as required string with 1-200 character length constraint",
|
|
"keywords": ["name", "required", "string", "length", "200", "validate"],
|
|
"explanation": "The name field must be validated as a required string with length between 1 and 200 characters."
|
|
},
|
|
{
|
|
"id": "IMPL-IV-2",
|
|
"severity": "CRITICAL",
|
|
"category": "finding",
|
|
"summary": "Must validate price as required non-negative number with max 2 decimal places",
|
|
"keywords": ["price", "number", "non-negative", "decimal", "places", "validate"],
|
|
"explanation": "Price must be >= 0 and have at most 2 decimal places. This prevents values like -5 or 19.999."
|
|
},
|
|
{
|
|
"id": "IMPL-IV-3",
|
|
"severity": "CRITICAL",
|
|
"category": "finding",
|
|
"summary": "Must validate SKU against pattern ^[A-Z]{2,4}-\\d{4,8}$ — alphanumeric prefix with numeric suffix",
|
|
"keywords": ["SKU", "pattern", "regex", "validate", "format"],
|
|
"explanation": "SKU must match the specific pattern: 2-4 uppercase letters, a dash, then 4-8 digits."
|
|
},
|
|
{
|
|
"id": "IMPL-IV-4",
|
|
"severity": "MAJOR",
|
|
"category": "finding",
|
|
"summary": "Must validate category against enum — only electronics, clothing, food, or other allowed",
|
|
"keywords": ["category", "enum", "valid", "electronics", "clothing", "food"],
|
|
"explanation": "Category must be one of the predefined values from the Product type."
|
|
},
|
|
{
|
|
"id": "IMPL-IV-5",
|
|
"severity": "MAJOR",
|
|
"category": "finding",
|
|
"summary": "Must return 400 status with descriptive error messages — not 500",
|
|
"keywords": ["400", "error", "message", "descriptive", "status", "validation"],
|
|
"explanation": "Validation failures should return HTTP 400 with clear error messages indicating which field failed and why."
|
|
},
|
|
{
|
|
"id": "IMPL-IV-6",
|
|
"severity": "MAJOR",
|
|
"category": "finding",
|
|
"summary": "Must not modify the Product interface or existing GET route — validation is additive only",
|
|
"keywords": ["modify", "Product", "interface", "GET", "route", "existing"],
|
|
"explanation": "The task explicitly states not to modify the Product interface or existing GET route. Validation should be added as middleware or inline in the POST handler."
|
|
}
|
|
]
|
|
}
|