feat: implement FlowExecutor for executing flow definitions with DAG traversal and node execution

This commit is contained in:
catlog22
2026-01-30 16:59:18 +08:00
parent 0a7c1454d9
commit a5c3dff8d3
92 changed files with 23875 additions and 542 deletions

File diff suppressed because it is too large Load Diff

View File

@@ -0,0 +1,391 @@
---
name: code-validation-gate
description: Validate AI-generated code for common errors (imports, variables, types) before test execution
argument-hint: "--session WFS-test-session-id [--fix] [--strict]"
examples:
- /workflow:tools:code-validation-gate --session WFS-test-auth
- /workflow:tools:code-validation-gate --session WFS-test-auth --fix
- /workflow:tools:code-validation-gate --session WFS-test-auth --strict
---
# Code Validation Gate Command
## Overview
Pre-test validation gate that checks AI-generated code for common errors before test execution. This prevents wasted test cycles on code with fundamental issues like import errors, variable conflicts, and type mismatches.
## Core Philosophy
- **Fail Fast**: Catch fundamental errors before expensive test execution
- **AI-Aware**: Specifically targets common AI code generation mistakes
- **Auto-Remediation**: Attempt safe fixes before failing
- **Clear Feedback**: Provide actionable fix suggestions for manual intervention
## Target Error Categories
### L0.1: Compilation Errors
- TypeScript compilation failures
- Syntax errors
- Module resolution failures
### L0.2: Import Errors
- Unresolved module imports (hallucinated packages)
- Circular dependencies
- Duplicate imports
- Unused imports
### L0.3: Variable Errors
- Variable redeclaration
- Scope conflicts (shadowing)
- Undefined variable usage
- Unused variables
### L0.4: Type Errors (TypeScript)
- Type mismatches
- Missing type definitions
- Excessive `any` usage
- Implicit `any` types
### L0.5: AI-Specific Patterns
- Placeholder code (`// TODO: implement`)
- Hallucinated package imports
- Mock code in production files
- Inconsistent naming patterns
## Execution Process
```
Input Parsing:
├─ Parse flags: --session (required), --fix, --strict
└─ Load test-quality-config.json
Phase 1: Context Loading
├─ Load session metadata
├─ Identify target files (from IMPL-001 output or context-package)
└─ Detect project configuration (tsconfig, eslint, etc.)
Phase 2: Validation Execution
├─ L0.1: Run TypeScript compilation check
├─ L0.2: Run import validation
├─ L0.3: Run variable validation
├─ L0.4: Run type validation
└─ L0.5: Run AI-specific checks
Phase 3: Result Analysis
├─ Aggregate all findings by severity
├─ Calculate pass/fail status
└─ Generate fix suggestions
Phase 4: Auto-Fix (if --fix enabled)
├─ Apply safe auto-fixes (imports, formatting)
├─ Re-run validation
└─ Report remaining issues
Phase 5: Gate Decision
├─ PASS: Proceed to IMPL-001.5
├─ SOFT_FAIL: Auto-fix applied, needs re-validation
└─ HARD_FAIL: Block with detailed report
```
## Execution Lifecycle
### Phase 1: Context Loading
**Load session and identify validation targets.**
```javascript
// Load session metadata
Read(".workflow/active/{session_id}/workflow-session.json")
// Load context package for target files
Read(".workflow/active/{session_id}/.process/context-package.json")
// OR
Read(".workflow/active/{session_id}/.process/test-context-package.json")
// Identify files to validate:
// 1. Source files from context.implementation_files
// 2. Test files from IMPL-001 output (if exists)
// 3. All modified files since session start
```
**Target File Discovery**:
- Source files: `context.focus_paths` from context-package
- Generated tests: `.workflow/active/{session_id}/.task/IMPL-001-output/`
- All TypeScript/JavaScript in target directories
### Phase 2: Validation Execution
**Execute validation checks in order of dependency.**
#### L0.1: TypeScript Compilation
```bash
# Primary check - catches most fundamental errors
npx tsc --noEmit --skipLibCheck --project tsconfig.json 2>&1
# Parse output for errors
# Critical: Any compilation error blocks further validation
```
**Error Patterns**:
```
error TS2307: Cannot find module 'xxx'
error TS2451: Cannot redeclare block-scoped variable 'xxx'
error TS2322: Type 'xxx' is not assignable to type 'yyy'
```
#### L0.2: Import Validation
```bash
# Check for circular dependencies
npx madge --circular --extensions ts,tsx,js,jsx {target_dirs}
# ESLint import rules
npx eslint --rule 'import/no-duplicates: error' --rule 'import/no-unresolved: error' {files}
```
**Hallucinated Package Check**:
```javascript
// Extract all imports from files
// Verify each package exists in package.json or node_modules
// Flag any unresolvable imports as "hallucinated"
```
#### L0.3: Variable Validation
```bash
# ESLint variable rules
npx eslint --rule 'no-shadow: error' --rule 'no-undef: error' --rule 'no-redeclare: error' {files}
```
#### L0.4: Type Validation
```bash
# TypeScript strict checks
npx tsc --noEmit --strict {files}
# Check for any abuse
npx eslint --rule '@typescript-eslint/no-explicit-any: warn' {files}
```
#### L0.5: AI-Specific Checks
```bash
# Check for placeholder code
grep -rn "// TODO: implement\|// Add your code here\|throw new Error.*Not implemented" {files}
# Check for mock code in production files
grep -rn "jest\.mock\|sinon\.\|vi\.mock" {source_files_only}
```
### Phase 3: Result Analysis
**Aggregate and categorize findings.**
```javascript
const findings = {
critical: [], // Blocks all progress
error: [], // Blocks with threshold
warning: [] // Advisory only
};
// Apply thresholds from config
const config = loadConfig("test-quality-config.json");
const thresholds = config.code_validation.severity_thresholds;
// Gate decision
if (findings.critical.length > thresholds.critical) {
decision = "HARD_FAIL";
} else if (findings.error.length > thresholds.error) {
decision = "SOFT_FAIL"; // Try auto-fix
} else {
decision = "PASS";
}
```
### Phase 4: Auto-Fix (Optional)
**Apply safe automatic fixes when --fix flag provided.**
```bash
# Safe fixes only
npx eslint --fix --rule 'import/no-duplicates: error' --rule 'unused-imports/no-unused-imports: error' {files}
# Re-run validation after fixes
# Report what was fixed vs what remains
```
**Safe Fix Categories**:
- Remove unused imports
- Remove duplicate imports
- Fix import ordering
- Remove unused variables (with caution)
- Formatting fixes
**Unsafe (Manual Only)**:
- Missing imports (need to determine correct package)
- Type errors (need to understand intent)
- Variable shadowing (need to understand scope intent)
### Phase 5: Gate Decision
**Determine next action based on results.**
| Decision | Condition | Action |
|----------|-----------|--------|
| **PASS** | critical=0, error<=3, warning<=10 | Proceed to IMPL-001.5 |
| **SOFT_FAIL** | critical=0, error>3 OR fixable issues | Auto-fix and retry (max 2) |
| **HARD_FAIL** | critical>0 OR max retries exceeded | Block with report |
## Output Artifacts
### Validation Report
**File**: `.workflow/active/{session_id}/.process/code-validation-report.md`
```markdown
# Code Validation Report
**Session**: {session_id}
**Timestamp**: {timestamp}
**Status**: PASS | SOFT_FAIL | HARD_FAIL
## Summary
- Files Validated: {count}
- Critical Issues: {count}
- Errors: {count}
- Warnings: {count}
## Critical Issues (Must Fix)
### Import Errors
- `src/auth/service.ts:5` - Cannot find module 'non-existent-package'
- **Suggestion**: Check if package exists, may be hallucinated by AI
### Variable Conflicts
- `src/utils/helper.ts:12` - Cannot redeclare block-scoped variable 'config'
- **Suggestion**: Rename one of the variables or merge declarations
## Errors (Should Fix)
...
## Warnings (Consider Fixing)
...
## Auto-Fix Applied
- Removed 3 unused imports in `src/auth/service.ts`
- Fixed import ordering in `src/utils/index.ts`
## Remaining Issues Requiring Manual Fix
...
## Next Steps
- [ ] Fix critical issues before proceeding
- [ ] Review error suggestions
- [ ] Re-run validation: `/workflow:tools:code-validation-gate --session {session_id}`
```
### JSON Report (Machine-Readable)
**File**: `.workflow/active/{session_id}/.process/code-validation-report.json`
```json
{
"session_id": "WFS-test-xxx",
"timestamp": "2025-01-30T10:00:00Z",
"status": "HARD_FAIL",
"summary": {
"files_validated": 15,
"critical": 2,
"error": 5,
"warning": 8
},
"findings": {
"critical": [
{
"category": "import",
"file": "src/auth/service.ts",
"line": 5,
"message": "Cannot find module 'non-existent-package'",
"suggestion": "Check if package exists in package.json",
"auto_fixable": false
}
],
"error": [...],
"warning": [...]
},
"auto_fixes_applied": [...],
"gate_decision": "HARD_FAIL",
"retry_count": 0,
"max_retries": 2
}
```
## Command Options
| Option | Description | Default |
|--------|-------------|---------|
| `--session` | Test session ID (required) | - |
| `--fix` | Enable auto-fix for safe issues | false |
| `--strict` | Use strict thresholds (0 errors allowed) | false |
| `--files` | Specific files to validate (comma-separated) | All target files |
| `--skip-types` | Skip TypeScript type checks | false |
## Integration
### Command Chain
- **Called By**: `/workflow:test-fix-gen` (after IMPL-001)
- **Requires**: IMPL-001 output OR context-package.json
- **Followed By**: IMPL-001.5 (Test Quality Gate) on PASS
### Task JSON Integration
When used in test-fix workflow, generates task:
```json
{
"id": "IMPL-001.3-validation",
"meta": {
"type": "code-validation",
"agent": "@test-fix-agent"
},
"context": {
"depends_on": ["IMPL-001"],
"requirements": "Validate generated code for AI common errors"
},
"flow_control": {
"validation_config": "~/.claude/workflows/test-quality-config.json",
"max_retries": 2,
"auto_fix_enabled": true
},
"acceptance_criteria": [
"Zero critical issues",
"Maximum 3 error issues",
"All imports resolvable",
"No variable redeclarations"
]
}
```
## Error Handling
| Error | Resolution |
|-------|------------|
| tsconfig.json not found | Use default compiler options |
| ESLint not installed | Skip ESLint checks, use tsc only |
| madge not installed | Skip circular dependency check |
| No files to validate | Return PASS (nothing to check) |
## Best Practices
1. **Run Early**: Execute validation immediately after code generation
2. **Use --fix First**: Let auto-fix resolve trivial issues
3. **Review Suggestions**: AI fix suggestions may need human judgment
4. **Don't Skip Critical**: Never proceed with critical errors
5. **Track Patterns**: Common failures indicate prompt improvement opportunities
## Related Commands
- `/workflow:test-fix-gen` - Parent workflow that invokes this command
- `/workflow:tools:test-quality-gate` - Next phase (IMPL-001.5) for test quality
- `/workflow:test-cycle-execute` - Execute tests after validation passes

View File

@@ -143,7 +143,7 @@ Determine CLI tool usage per-step based on user's task description:
(Detailed specifications in your agent definition)
### Task Structure Requirements
- Minimum 2 tasks: IMPL-001 (test generation) + IMPL-002 (test execution & fix)
- Minimum 4 tasks: IMPL-001 (test generation) + IMPL-001.3 (code validation) + IMPL-001.5 (test quality) + IMPL-002 (test execution & fix)
- Expandable for complex projects: Add IMPL-003+ (per-module, integration, E2E tests)
Task Configuration:
@@ -154,9 +154,29 @@ Task Configuration:
- flow_control: Test generation strategy from TEST_ANALYSIS_RESULTS.md
- CLI execution: Add `command` field when user requests (determined semantically)
IMPL-001.3 (Code Validation Gate) ← NEW:
- meta.type: "code-validation"
- meta.agent: "@test-fix-agent"
- context.depends_on: ["IMPL-001"]
- context.validation_config: "~/.claude/workflows/test-quality-config.json"
- flow_control.validation_phases: ["compilation", "imports", "variables", "types", "ai_specific"]
- flow_control.auto_fix_enabled: true
- flow_control.max_retries: 2
- flow_control.severity_thresholds: { critical: 0, error: 3, warning: 10 }
- acceptance_criteria: Zero compilation errors, all imports resolvable, no variable redeclarations
IMPL-001.5 (Test Quality Gate):
- meta.type: "test-quality-review"
- meta.agent: "@test-fix-agent"
- context.depends_on: ["IMPL-001", "IMPL-001.3"]
- context.quality_config: "~/.claude/workflows/test-quality-config.json"
- flow_control: Static analysis, coverage analysis, anti-pattern detection
- acceptance_criteria: Coverage ≥ 80%, zero critical anti-patterns
IMPL-002+ (Test Execution & Fix):
- meta.type: "test-fix"
- meta.agent: "@test-fix-agent"
- context.depends_on: ["IMPL-001", "IMPL-001.3", "IMPL-001.5"]
- flow_control: Test-fix cycle with iteration limits and diagnosis configuration
- CLI execution: Add `command` field when user requests (determined semantically)
@@ -190,10 +210,17 @@ PRIMARY requirements source - extract and map to task JSONs:
- Implementation targets → context.files_to_test (absolute paths)
## EXPECTED DELIVERABLES
1. Test Task JSON Files (.task/IMPL-*.json)
1. Test Task JSON Files (.task/IMPL-*.json) - Minimum 4 required:
- IMPL-001.json: Test generation task
- IMPL-001.3-validation.json: Code validation gate (AI error detection) ← NEW
- IMPL-001.5-review.json: Test quality gate
- IMPL-002.json: Test execution & fix cycle
Each task includes:
- 6-field schema with quantified requirements from TEST_ANALYSIS_RESULTS.md
- Test-specific metadata: type, agent, test_framework, coverage_target
- flow_control includes: reusable_test_tools, test_commands (from project config)
- Validation config reference for IMPL-001.3: ~/.claude/workflows/test-quality-config.json
- CLI execution via `command` field when user requests (determined semantically)
- Artifact references from test-context-package.json
- Absolute paths in context.files_to_test
@@ -211,7 +238,7 @@ PRIMARY requirements source - extract and map to task JSONs:
## QUALITY STANDARDS
Hard Constraints:
- Task count: minimum 2, maximum 18
- Task count: minimum 4, maximum 18 (IMPL-001, IMPL-001.3, IMPL-001.5, IMPL-002 required)
- All requirements quantified from TEST_ANALYSIS_RESULTS.md
- Test framework matches existing project framework
- flow_control includes reusable_test_tools and test_commands from project
@@ -249,7 +276,11 @@ CLI tool usage is determined semantically from user's task description:
- Default: Agent execution (no `command` field)
### Output
- Test task JSON files in `.task/` directory (minimum 2)
- IMPL_PLAN.md with test strategy and fix cycle specification
- Test task JSON files in `.task/` directory (minimum 4):
- IMPL-001.json (test generation)
- IMPL-001.3-validation.json (code validation gate)
- IMPL-001.5-review.json (test quality gate)
- IMPL-002.json (test execution & fix)
- IMPL_PLAN.md with test strategy, validation gates, and fix cycle specification
- TODO_LIST.md with test phase indicators
- Session ready for test execution

View File

@@ -0,0 +1,251 @@
{
"$schema": "https://json-schema.org/draft/2020-12/schema",
"version": "1.0.0",
"description": "Test quality and code validation configuration for AI-generated code",
"code_validation": {
"description": "Pre-test validation for AI-generated code common errors",
"enabled": true,
"phases": {
"L0_compilation": {
"description": "TypeScript/JavaScript compilation check",
"enabled": true,
"commands": {
"typescript": "npx tsc --noEmit --skipLibCheck",
"javascript": "node --check"
},
"critical": true,
"failure_blocks_tests": true
},
"L0_imports": {
"description": "Import statement validation",
"enabled": true,
"checks": [
{
"id": "unresolved_imports",
"description": "Check for unresolved module imports",
"pattern": "Cannot find module|Module not found|Unable to resolve",
"severity": "critical"
},
{
"id": "circular_imports",
"description": "Check for circular dependencies",
"tool": "madge",
"command": "npx madge --circular --extensions ts,tsx,js,jsx",
"severity": "warning"
},
{
"id": "duplicate_imports",
"description": "Check for duplicate imports",
"eslint_rule": "import/no-duplicates",
"severity": "error"
},
{
"id": "unused_imports",
"description": "Check for unused imports",
"eslint_rule": "unused-imports/no-unused-imports",
"severity": "warning"
}
]
},
"L0_variables": {
"description": "Variable declaration validation",
"enabled": true,
"checks": [
{
"id": "redeclaration",
"description": "Check for variable redeclaration",
"pattern": "Cannot redeclare|Duplicate identifier|has already been declared",
"severity": "critical"
},
{
"id": "scope_conflict",
"description": "Check for scope conflicts",
"eslint_rule": "no-shadow",
"severity": "error"
},
{
"id": "undefined_vars",
"description": "Check for undefined variables",
"eslint_rule": "no-undef",
"severity": "critical"
},
{
"id": "unused_vars",
"description": "Check for unused variables",
"eslint_rule": "@typescript-eslint/no-unused-vars",
"severity": "warning"
}
]
},
"L0_types": {
"description": "TypeScript type validation",
"enabled": true,
"checks": [
{
"id": "type_mismatch",
"description": "Check for type mismatches",
"pattern": "Type .* is not assignable to type",
"severity": "critical"
},
{
"id": "missing_types",
"description": "Check for missing type definitions",
"pattern": "Could not find a declaration file",
"severity": "warning"
},
{
"id": "any_abuse",
"description": "Check for excessive any type usage",
"eslint_rule": "@typescript-eslint/no-explicit-any",
"severity": "warning",
"max_occurrences": 5
},
{
"id": "implicit_any",
"description": "Check for implicit any",
"pattern": "implicitly has an 'any' type",
"severity": "error"
}
]
}
},
"severity_thresholds": {
"critical": 0,
"error": 3,
"warning": 10
},
"max_retries": 2,
"auto_fix": {
"enabled": true,
"safe_fixes_only": true,
"fixable_categories": ["imports", "formatting", "unused_vars"]
}
},
"test_quality": {
"description": "Test file quality validation (IMPL-001.5)",
"enabled": true,
"coverage": {
"minimum_threshold": 80,
"branch_threshold": 70,
"function_threshold": 80,
"line_threshold": 80
},
"anti_patterns": {
"empty_test_body": {
"pattern": "it\\(['\"].*['\"],\\s*\\(\\)\\s*=>\\s*\\{\\s*\\}\\)",
"severity": "critical",
"description": "Test with empty body"
},
"missing_assertion": {
"pattern": "it\\(['\"].*['\"],.*\\{[^}]*\\}\\)(?![\\s\\S]*expect)",
"severity": "critical",
"description": "Test without expect() assertion"
},
"skipped_without_reason": {
"pattern": "(it|describe)\\.skip\\(['\"][^'\"]*['\"](?!.*\\/\\/ )",
"severity": "error",
"description": "Skipped test without comment explaining why"
},
"todo_test": {
"pattern": "(it|test)\\.todo\\(",
"severity": "warning",
"description": "TODO test placeholder"
},
"only_test": {
"pattern": "(it|describe)\\.only\\(",
"severity": "critical",
"description": "Focused test (will skip other tests)"
}
},
"required_test_types": {
"unit": {
"min_per_function": 1,
"must_include": ["happy_path"]
},
"negative": {
"min_per_public_api": 1,
"description": "Error handling tests for public APIs"
},
"edge_case": {
"required_scenarios": ["null", "undefined", "empty_string", "empty_array", "boundary_values"]
}
}
},
"ai_specific_checks": {
"description": "Checks specifically for AI-generated code patterns",
"enabled": true,
"checks": [
{
"id": "hallucinated_imports",
"description": "Check for imports of non-existent packages",
"validation": "npm_package_exists",
"severity": "critical"
},
{
"id": "inconsistent_naming",
"description": "Check for naming inconsistencies within file",
"pattern": "function (\\w+).*\\1(?!\\()",
"severity": "warning"
},
{
"id": "placeholder_code",
"description": "Check for AI placeholder comments",
"patterns": [
"// TODO: implement",
"// Add your code here",
"// Implementation pending",
"throw new Error\\(['\"]Not implemented['\"]\\)"
],
"severity": "error"
},
{
"id": "mock_in_production",
"description": "Check for mock/stub code in production files",
"patterns": [
"jest\\.mock\\(",
"sinon\\.",
"vi\\.mock\\("
],
"exclude_paths": ["**/*.test.*", "**/*.spec.*", "**/test/**", "**/__tests__/**"],
"severity": "critical"
}
]
},
"validation_commands": {
"typescript_check": {
"command": "npx tsc --noEmit --skipLibCheck",
"timeout": 60000,
"parse_errors": true
},
"eslint_check": {
"command": "npx eslint --format json",
"timeout": 60000,
"auto_fix_command": "npx eslint --fix"
},
"circular_deps_check": {
"command": "npx madge --circular --extensions ts,tsx,js,jsx",
"timeout": 30000
},
"package_validation": {
"command": "npm ls --json",
"timeout": 30000
}
},
"gate_decisions": {
"pass_criteria": {
"critical_issues": 0,
"error_issues": "<=3",
"warning_issues": "<=10"
},
"actions": {
"pass": "Proceed to IMPL-001.5 (Test Quality Gate)",
"soft_fail": "Auto-fix and retry (max 2 attempts)",
"hard_fail": "Block and report to user with fix suggestions"
}
}
}