agentic-qe 3.4.1 → 3.4.2

package/.claude/skills/sherlock-review/SKILL.md

@@ -11,6 +11,11 @@ last_optimized: 2025-12-03
 dependencies: []
 quick_reference_card: true
 tags: [investigation, evidence-based, code-review, root-cause, deduction]
+trust_tier: 2
+validation:
+  schema_path: schemas/output.json
+  validator_path: scripts/validate.sh
+
 ---
 
 # Sherlock Review

package/.claude/skills/sherlock-review/schemas/output.json

@@ -0,0 +1,297 @@
+{
+  "$schema": "https://json-schema.org/draft/2020-12/schema",
+  "$id": "https://agentic-qe.dev/schemas/skills/sherlock-review/output.json",
+  "title": "Sherlock Review Skill Output Schema",
+  "description": "Schema for sherlock-review skill output. Evidence-based investigative code review with deductive reasoning.",
+  "type": "object",
+  "required": ["skillName", "version", "timestamp", "status", "trustTier", "output"],
+  "properties": {
+    "skillName": {
+      "type": "string",
+      "const": "sherlock-review"
+    },
+    "version": {
+      "type": "string",
+      "pattern": "^\\d+\\.\\d+\\.\\d+(-[a-zA-Z0-9]+)?$"
+    },
+    "timestamp": {
+      "type": "string",
+      "format": "date-time"
+    },
+    "status": {
+      "type": "string",
+      "enum": ["success", "partial", "failed", "skipped"]
+    },
+    "trustTier": {
+      "type": "integer",
+      "const": 3
+    },
+    "output": {
+      "type": "object",
+      "required": ["summary", "reviewFindings", "deductions", "codeQuality"],
+      "properties": {
+        "summary": {
+          "type": "string",
+          "minLength": 50,
+          "maxLength": 3000,
+          "description": "Investigation summary with key deductions"
+        },
+        "reviewFindings": {
+          "type": "array",
+          "items": {
+            "$ref": "#/$defs/reviewFinding"
+          },
+          "maxItems": 200
+        },
+        "deductions": {
+          "type": "array",
+          "items": {
+            "$ref": "#/$defs/deduction"
+          },
+          "description": "Logical deductions based on evidence"
+        },
+        "codeQuality": {
+          "$ref": "#/$defs/codeQuality"
+        },
+        "severity": {
+          "$ref": "#/$defs/severityBreakdown"
+        },
+        "recommendations": {
+          "type": "array",
+          "items": {
+            "$ref": "#/$defs/recommendation"
+          },
+          "maxItems": 100
+        },
+        "evidenceChain": {
+          "type": "array",
+          "items": {
+            "$ref": "#/$defs/evidence"
+          },
+          "description": "Chain of evidence supporting findings"
+        },
+        "suspiciousPatterns": {
+          "type": "array",
+          "items": {
+            "$ref": "#/$defs/suspiciousPattern"
+          }
+        },
+        "verdict": {
+          "type": "string",
+          "enum": ["case-closed-clean", "minor-infractions", "significant-concerns", "major-violations", "investigation-required"]
+        },
+        "metrics": {
+          "$ref": "#/$defs/metrics"
+        }
+      }
+    },
+    "metadata": {
+      "$ref": "#/$defs/metadata"
+    },
+    "validation": {
+      "$ref": "#/$defs/validationResult"
+    },
+    "learning": {
+      "$ref": "#/$defs/learningData"
+    }
+  },
+  "$defs": {
+    "reviewFinding": {
+      "type": "object",
+      "required": ["id", "title", "severity", "category", "evidence"],
+      "properties": {
+        "id": {
+          "type": "string",
+          "pattern": "^SHRK-\\d{3,6}$"
+        },
+        "title": {
+          "type": "string",
+          "minLength": 10,
+          "maxLength": 200
+        },
+        "description": {
+          "type": "string",
+          "maxLength": 2000
+        },
+        "severity": {
+          "type": "string",
+          "enum": ["critical", "high", "medium", "low", "observation"]
+        },
+        "category": {
+          "type": "string",
+          "enum": ["logic-flaw", "security-risk", "performance-issue", "design-smell", "hidden-bug", "race-condition", "data-integrity", "edge-case", "assumption-violation"]
+        },
+        "evidence": {
+          "type": "string",
+          "minLength": 20,
+          "maxLength": 3000,
+          "description": "Evidence supporting this finding"
+        },
+        "deductiveReasoning": {
+          "type": "string",
+          "maxLength": 2000,
+          "description": "Step-by-step deductive reasoning"
+        },
+        "location": {
+          "type": "object",
+          "properties": {
+            "file": { "type": "string" },
+            "line": { "type": "integer", "minimum": 1 },
+            "endLine": { "type": "integer", "minimum": 1 },
+            "function": { "type": "string" }
+          }
+        },
+        "codeSnippet": {
+          "type": "string",
+          "maxLength": 5000
+        },
+        "suggestedFix": {
+          "type": "string",
+          "maxLength": 5000
+        },
+        "confidence": {
+          "type": "number",
+          "minimum": 0,
+          "maximum": 1,
+          "description": "Confidence in this deduction"
+        }
+      }
+    },
+    "deduction": {
+      "type": "object",
+      "required": ["id", "observation", "hypothesis", "conclusion"],
+      "properties": {
+        "id": {
+          "type": "string",
+          "pattern": "^DED-\\d{3,6}$"
+        },
+        "observation": {
+          "type": "string",
+          "description": "What was observed in the code"
+        },
+        "hypothesis": {
+          "type": "string",
+          "description": "What this might indicate"
+        },
+        "evidence": {
+          "type": "array",
+          "items": { "type": "string" },
+          "description": "Supporting evidence"
+        },
+        "conclusion": {
+          "type": "string",
+          "description": "Deductive conclusion"
+        },
+        "confidence": {
+          "type": "number",
+          "minimum": 0,
+          "maximum": 1
+        },
+        "relatedFindings": {
+          "type": "array",
+          "items": { "type": "string" }
+        }
+      }
+    },
+    "evidence": {
+      "type": "object",
+      "required": ["type", "description"],
+      "properties": {
+        "type": {
+          "type": "string",
+          "enum": ["code-pattern", "data-flow", "control-flow", "state-mutation", "external-call", "configuration", "comment", "commit-history"]
+        },
+        "description": { "type": "string" },
+        "location": { "type": "string" },
+        "significance": { "type": "string", "enum": ["critical", "supporting", "circumstantial"] }
+      }
+    },
+    "suspiciousPattern": {
+      "type": "object",
+      "required": ["patternType", "description", "locations"],
+      "properties": {
+        "patternType": {
+          "type": "string",
+          "enum": ["anti-pattern", "code-smell", "security-risk", "performance-trap", "maintenance-nightmare", "time-bomb"]
+        },
+        "description": { "type": "string" },
+        "locations": { "type": "array", "items": { "type": "string" } },
+        "riskLevel": { "type": "string", "enum": ["critical", "high", "medium", "low"] }
+      }
+    },
+    "codeQuality": {
+      "type": "object",
+      "required": ["overallScore", "grade"],
+      "properties": {
+        "overallScore": { "type": "number", "minimum": 0, "maximum": 100 },
+        "grade": { "type": "string", "pattern": "^[A-F][+-]?$" },
+        "dimensions": {
+          "type": "object",
+          "properties": {
+            "correctness": { "type": "number", "minimum": 0, "maximum": 100 },
+            "clarity": { "type": "number", "minimum": 0, "maximum": 100 },
+            "robustness": { "type": "number", "minimum": 0, "maximum": 100 },
+            "security": { "type": "number", "minimum": 0, "maximum": 100 },
+            "testability": { "type": "number", "minimum": 0, "maximum": 100 }
+          }
+        }
+      }
+    },
+    "severityBreakdown": {
+      "type": "object",
+      "properties": {
+        "critical": { "type": "integer", "minimum": 0 },
+        "high": { "type": "integer", "minimum": 0 },
+        "medium": { "type": "integer", "minimum": 0 },
+        "low": { "type": "integer", "minimum": 0 },
+        "observation": { "type": "integer", "minimum": 0 }
+      }
+    },
+    "recommendation": {
+      "type": "object",
+      "required": ["id", "title", "priority"],
+      "properties": {
+        "id": { "type": "string", "pattern": "^REC-\\d{3,6}$" },
+        "title": { "type": "string", "maxLength": 200 },
+        "description": { "type": "string", "maxLength": 2000 },
+        "priority": { "type": "string", "enum": ["critical", "high", "medium", "low"] },
+        "evidence": { "type": "string" },
+        "codeExample": { "type": "string", "maxLength": 5000 }
+      }
+    },
+    "metrics": {
+      "type": "object",
+      "properties": {
+        "filesInvestigated": { "type": "integer", "minimum": 0 },
+        "linesAnalyzed": { "type": "integer", "minimum": 0 },
+        "deductionsMade": { "type": "integer", "minimum": 0 },
+        "suspiciousPatternsFound": { "type": "integer", "minimum": 0 },
+        "evidenceCollected": { "type": "integer", "minimum": 0 }
+      }
+    },
+    "metadata": {
+      "type": "object",
+      "properties": {
+        "executionTimeMs": { "type": "integer", "minimum": 0 },
+        "toolsUsed": { "type": "array", "items": { "type": "string" } },
+        "agentId": { "type": "string", "pattern": "^qe-[a-z][a-z0-9-]*$" },
+        "investigationType": { "type": "string", "enum": ["comprehensive", "focused", "targeted"] }
+      }
+    },
+    "validationResult": {
+      "type": "object",
+      "properties": {
+        "schemaValid": { "type": "boolean" },
+        "contentValid": { "type": "boolean" },
+        "confidence": { "type": "number", "minimum": 0, "maximum": 1 }
+      }
+    },
+    "learningData": {
+      "type": "object",
+      "properties": {
+        "patternsDetected": { "type": "array", "items": { "type": "string" } },
+        "reward": { "type": "number", "minimum": 0, "maximum": 1 }
+      }
+    }
+  }
+}

package/.claude/skills/sherlock-review/scripts/validate.sh

@@ -0,0 +1,115 @@
+#!/bin/bash
+# =============================================================================
+# AQE Skill Validator: sherlock-review v1.0.0
+# =============================================================================
+set -euo pipefail
+
+SCRIPT_DIR="$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)"
+SKILL_DIR="$(cd "$SCRIPT_DIR/.." && pwd)"
+PROJECT_ROOT="$(cd "$SKILL_DIR/../../.." && pwd)"
+
+for lib_path in "$PROJECT_ROOT/.claude/skills/.validation/templates/validator-lib.sh" "$SKILL_DIR/scripts/validator-lib.sh"; do
+  [[ -f "$lib_path" ]] && source "$lib_path" && break
+done
+
+SKILL_NAME="sherlock-review"
+SKILL_VERSION="1.0.0"
+REQUIRED_TOOLS=("jq")
+SCHEMA_PATH="$SKILL_DIR/schemas/output.json"
+REQUIRED_FIELDS=("skillName" "status" "output" "output.summary" "output.reviewFindings" "output.deductions" "output.codeQuality")
+MUST_CONTAIN_TERMS=("evidence" "deduction")
+MUST_NOT_CONTAIN_TERMS=("TODO" "FIXME")
+ENUM_VALIDATIONS=(".status:success,partial,failed,skipped")
+
+validate_skill_specific() {
+  local output_file="$1"
+  local has_errors=false
+
+  # Validate review findings with evidence
+  local finding_count
+  finding_count=$(jq '.output.reviewFindings | length' "$output_file" 2>/dev/null || echo "0")
+
+  if [[ "$finding_count" -gt 0 ]]; then
+    local invalid_findings
+    invalid_findings=$(jq '[.output.reviewFindings[]? | select(.id == null or .title == null or .severity == null or .category == null or .evidence == null)] | length' "$output_file" 2>/dev/null || echo "0")
+
+    if [[ "$invalid_findings" -gt 0 ]]; then
+      warn "$invalid_findings finding(s) missing required fields (id, title, severity, category, evidence)"
+    fi
+
+    # Validate finding ID pattern
+    local invalid_ids
+    invalid_ids=$(jq '[.output.reviewFindings[]?.id // empty | select(test("^SHRK-[0-9]{3,6}$") | not)] | length' "$output_file" 2>/dev/null || echo "0")
+
+    if [[ "$invalid_ids" -gt 0 ]]; then
+      warn "$invalid_ids finding(s) have invalid ID format (should be SHRK-NNN)"
+    fi
+  fi
+
+  # Validate deductions - this is critical for Sherlock review
+  local deduction_count
+  deduction_count=$(jq '.output.deductions | length' "$output_file" 2>/dev/null || echo "0")
+
+  if [[ "$deduction_count" -eq 0 ]]; then
+    warn "No deductions found - Sherlock review should include logical deductions"
+  else
+    local invalid_deductions
+    invalid_deductions=$(jq '[.output.deductions[]? | select(.id == null or .observation == null or .hypothesis == null or .conclusion == null)] | length' "$output_file" 2>/dev/null || echo "0")
+
+    if [[ "$invalid_deductions" -gt 0 ]]; then
+      warn "$invalid_deductions deduction(s) missing required fields (id, observation, hypothesis, conclusion)"
+    fi
+
+    # Validate deduction ID pattern
+    local invalid_ded_ids
+    invalid_ded_ids=$(jq '[.output.deductions[]?.id // empty | select(test("^DED-[0-9]{3,6}$") | not)] | length' "$output_file" 2>/dev/null || echo "0")
+
+    if [[ "$invalid_ded_ids" -gt 0 ]]; then
+      warn "$invalid_ded_ids deduction(s) have invalid ID format (should be DED-NNN)"
+    fi
+  fi
+
+  # Validate evidence chain if present
+  local evidence_count
+  evidence_count=$(jq '.output.evidenceChain | length' "$output_file" 2>/dev/null || echo "0")
+
+  if [[ "$evidence_count" -gt 0 ]]; then
+    local invalid_evidence
+    invalid_evidence=$(jq '[.output.evidenceChain[]? | select(.type == null or .description == null)] | length' "$output_file" 2>/dev/null || echo "0")
+
+    if [[ "$invalid_evidence" -gt 0 ]]; then
+      warn "$invalid_evidence evidence item(s) missing required fields"
+    fi
+  fi
+
+  # Validate verdict
+  local verdict
+  verdict=$(jq -r '.output.verdict // ""' "$output_file" 2>/dev/null)
+
+  if [[ -n "$verdict" && "$verdict" != "null" ]]; then
+    local valid_verdicts="case-closed-clean minor-infractions significant-concerns major-violations investigation-required"
+    if ! echo "$valid_verdicts" | grep -qw "$verdict"; then
+      error "Invalid verdict: $verdict"
+      has_errors=true
+    fi
+  fi
+
+  if [[ "$has_errors" == "true" ]]; then return 1; fi
+  success "Sherlock review validation passed"
+  return 0
+}
+
+OUTPUT_FILE="${1:-}"
+[[ -z "$OUTPUT_FILE" ]] && { echo "Usage: $0 <output-file>"; exit 1; }
+[[ ! -f "$OUTPUT_FILE" ]] && { echo "ERROR: File not found: $OUTPUT_FILE"; exit 1; }
+
+echo "Validating $SKILL_NAME output..."
+jq empty "$OUTPUT_FILE" 2>/dev/null || { echo "ERROR: Invalid JSON"; exit 1; }
+
+if validate_skill_specific "$OUTPUT_FILE"; then
+  echo "PASSED: $SKILL_NAME validation"
+  exit 0
+else
+  echo "FAILED: $SKILL_NAME validation"
+  exit 1
+fi

package/.claude/skills/shift-left-testing/SKILL.md

@@ -11,6 +11,12 @@ last_optimized: 2025-12-02
 dependencies: []
 quick_reference_card: true
 tags: [shift-left, early-testing, tdd, bdd, ci-cd, prevention]
+trust_tier: 3
+validation:
+  schema_path: schemas/output.json
+  validator_path: scripts/validate.sh
+  eval_path: evals/shift-left-testing.yaml
+
 ---
 
 # Shift-Left Testing

package/.claude/skills/shift-left-testing/evals/shift-left-testing.yaml

@@ -0,0 +1,145 @@
+skill: shift-left-testing
+version: 1.0.0
+description: >
+  Evaluation suite for shift-left testing strategy.
+  Tests early defect detection, pre-commit checks, and development-phase validation.
+
+models_to_test:
+  - claude-3.5-sonnet
+  - claude-3-haiku
+
+mcp_integration:
+  enabled: true
+  namespace: skill-validation
+  query_patterns: true
+  track_outcomes: true
+  store_patterns: true
+  target_agents:
+    - qe-learning-coordinator
+
+learning:
+  store_success_patterns: true
+  pattern_ttl_days: 90
+
+result_format:
+  json_output: true
+  include_timing: true
+  include_token_usage: true
+
+setup:
+  required_tools:
+    - jq
+
+test_cases:
+
+  - id: tc001_pre_commit_validation
+    description: "Identify issues before commit"
+    category: pre_commit
+    priority: critical
+
+    input:
+      code_changes:
+        - "function processPayment() { db.query(userInput); }"
+      checks:
+        - "sql_injection"
+        - "xss"
+        - "hardcoded_secrets"
+
+    expected_output:
+      must_contain:
+        - "SQL injection"
+        - "vulnerable"
+        - "pre-commit"
+
+    validation:
+      schema_check: true
+      keyword_match_threshold: 0.8
+
+  - id: tc002_unit_test_generation_early
+    description: "Generate unit tests during development"
+    category: development
+    priority: high
+
+    input:
+      code: |
+        function calculateDiscount(amount, percentage) {
+          return amount * (1 - percentage / 100);
+        }
+      generate_tests: true
+
+    expected_output:
+      must_contain:
+        - "test"
+        - "unit"
+        - "edge case"
+
+    validation:
+      schema_check: true
+
+  - id: tc003_code_quality_gates
+    description: "Apply quality gates during development"
+    category: quality
+    priority: high
+
+    input:
+      metrics:
+        complexity: 12
+        duplication_percent: 5
+        test_coverage: 75
+
+    expected_output:
+      must_contain:
+        - "quality"
+        - "gate"
+
+    validation:
+      schema_check: true
+
+  - id: tc004_early_architecture_review
+    description: "Validate architecture early in development"
+    category: architecture
+    priority: medium
+
+    input:
+      design:
+        layers: ["presentation", "business", "data"]
+        patterns: ["MVC", "dependency_injection"]
+
+    expected_output:
+      must_contain:
+        - "architecture"
+        - "design"
+
+    validation:
+      schema_check: true
+
+  - id: tc005_developer_feedback
+    description: "Provide immediate developer feedback"
+    category: feedback
+    priority: medium
+
+    input:
+      code_snippet: "const x = 1; const y = 1; const z = x + y;"
+      focus_areas: ["readability", "performance"]
+
+    expected_output:
+      must_contain:
+        - "feedback"
+        - "improve"
+
+    validation:
+      schema_check: true
+      allow_partial: true
+
+success_criteria:
+  pass_rate: 0.9
+  critical_pass_rate: 1.0
+  avg_reasoning_quality: 0.75
+  max_execution_time_ms: 300000
+
+metadata:
+  author: "qe-shift-left-coordinator"
+  created: "2026-02-02"
+  coverage_target: >
+    Shift-left testing with 5 test cases covering pre-commit validation,
+    early test generation, quality gates, architecture review, and developer feedback.