claude-flow-novice 2.6.0 → 2.8.1

package/.claude/skills/task-classifier/classify-task.sh

@@ -0,0 +1,115 @@
+#!/bin/bash
+
+# Task Type Classifier for CFN v3
+# Usage: classify-task.sh "task description"
+
+TASK_DESCRIPTION="${1:-}"
+
+if [ -z "$TASK_DESCRIPTION" ]; then
+  echo "Usage: classify-task.sh 'task description'" >&2
+  exit 1
+fi
+
+# Convert to lowercase for matching
+TASK_LOWER=$(echo "$TASK_DESCRIPTION" | tr '[:upper:]' '[:lower:]')
+
+# Keyword arrays per category
+SOFTWARE_KEYWORDS=(
+  "implement" "build" "code" "develop" "create api" "backend" "frontend"
+  "rest" "graphql" "database" "authentication" "bug fix" "refactor"
+  "optimize code" "web app" "full-stack" "api endpoint" "microservice"
+  "sdk" "library" "framework" "testing" "unit test" "integration test"
+)
+
+CONTENT_KEYWORDS=(
+  "write" "article" "blog post" "copy" "content" "documentation"
+  "guide" "tutorial" "whitepaper" "seo" "marketing copy"
+  "product description" "technical writing" "user manual" "readme"
+  "how-to" "explainer" "case study" "press release"
+)
+
+RESEARCH_KEYWORDS=(
+  "research" "analyze" "study" "investigate" "data analysis"
+  "statistical analysis" "literature review" "market research"
+  "competitive analysis" "feasibility study" "survey" "experiment"
+  "hypothesis" "methodology" "findings" "insights" "report"
+)
+
+DESIGN_KEYWORDS=(
+  "design" "ui" "ux" "mockup" "wireframe" "prototype"
+  "user interface" "user experience" "visual design" "branding"
+  "style guide" "accessibility" "responsive design" "figma"
+  "sketch" "adobe xd" "design system" "component library"
+)
+
+INFRASTRUCTURE_KEYWORDS=(
+  "deploy" "infrastructure" "devops" "cloud" "kubernetes" "docker"
+  "terraform" "aws" "azure" "gcp" "ci/cd" "monitoring" "logging"
+  "scaling" "networking" "security configuration" "load balancer"
+  "container" "orchestration" "helm" "ansible" "jenkins"
+)
+
+DATA_KEYWORDS=(
+  "etl" "pipeline" "data warehouse" "data lake" "streaming"
+  "real-time processing" "batch" "data quality" "schema"
+  "data model" "apache spark" "airflow" "data integration"
+  "olap" "oltp" "dimensional modeling" "star schema" "snowflake"
+  "databricks" "redshift" "bigquery" "data migration"
+)
+
+# Count matches per category
+count_matches() {
+  local -n keywords=$1
+  local count=0
+
+  for keyword in "${keywords[@]}"; do
+    if echo "$TASK_LOWER" | grep -q "$keyword"; then
+      ((count++))
+    fi
+  done
+
+  echo "$count"
+}
+
+SOFTWARE_COUNT=$(count_matches SOFTWARE_KEYWORDS)
+CONTENT_COUNT=$(count_matches CONTENT_KEYWORDS)
+RESEARCH_COUNT=$(count_matches RESEARCH_KEYWORDS)
+DESIGN_COUNT=$(count_matches DESIGN_KEYWORDS)
+INFRASTRUCTURE_COUNT=$(count_matches INFRASTRUCTURE_KEYWORDS)
+DATA_COUNT=$(count_matches DATA_KEYWORDS)
+
+# Find category with highest count
+MAX_COUNT=0
+TASK_TYPE="software-development"  # Default
+
+if [ "$SOFTWARE_COUNT" -gt "$MAX_COUNT" ]; then
+  MAX_COUNT=$SOFTWARE_COUNT
+  TASK_TYPE="software-development"
+fi
+
+if [ "$CONTENT_COUNT" -gt "$MAX_COUNT" ]; then
+  MAX_COUNT=$CONTENT_COUNT
+  TASK_TYPE="content-creation"
+fi
+
+if [ "$RESEARCH_COUNT" -gt "$MAX_COUNT" ]; then
+  MAX_COUNT=$RESEARCH_COUNT
+  TASK_TYPE="research"
+fi
+
+if [ "$DESIGN_COUNT" -gt "$MAX_COUNT" ]; then
+  MAX_COUNT=$DESIGN_COUNT
+  TASK_TYPE="design"
+fi
+
+if [ "$INFRASTRUCTURE_COUNT" -gt "$MAX_COUNT" ]; then
+  MAX_COUNT=$INFRASTRUCTURE_COUNT
+  TASK_TYPE="infrastructure"
+fi
+
+if [ "$DATA_COUNT" -gt "$MAX_COUNT" ]; then
+  MAX_COUNT=$DATA_COUNT
+  TASK_TYPE="data-engineering"
+fi
+
+echo "$TASK_TYPE"

package/.claude/skills/validation-templates/SKILL.md

@@ -0,0 +1,47 @@
+# Validation Templates Skill
+
+**Version:** 1.0.0
+**Purpose:** Domain-specific validation criteria for CFN Loop v3
+
+## Overview
+
+Provides predefined validation criteria for 6 task domains. Loop 2 validators use these to assess Loop 3 output quality.
+
+## Templates
+
+- `software.json` - Software development validation
+- `content.json` - Content creation validation
+- `research.json` - Research & analysis validation
+- `design.json` - Design & UX validation
+- `infrastructure.json` - Infrastructure & DevOps validation
+- `data.json` - Data engineering validation
+
+## Usage
+
+```bash
+# Load template for task type
+VALIDATION_CRITERIA=$(cat ./.claude/skills/validation-templates/software.json)
+
+# Extract critical criteria
+CRITICAL=$(echo "$VALIDATION_CRITERIA" | jq '.validation_criteria.critical[]')
+```
+
+## Criteria Levels
+
+**Critical** - Must pass for PROCEED decision
+- Blockers, security issues, core functionality
+- Examples: Tests pass, no vulnerabilities, build succeeds
+
+**Important** - Should pass but can iterate
+- Quality standards, coverage, documentation
+- Examples: Coverage ≥ 80%, linter clean, docs updated
+
+**Nice to Have** - Optional improvements
+- Optimizations, tech debt, enhancements
+- Examples: Performance improved, refactored code
+
+## Integration
+
+Used by:
+- `.claude/agents/cfn-v3-coordinator.md` - Load validation criteria
+- Loop 2 validator agents - Assess Loop 3 output

package/.claude/skills/validation-templates/content.json

@@ -0,0 +1,38 @@
+{
+  "domain": "content-creation",
+  "validation_criteria": {
+    "critical": [
+      "Content is original (no plagiarism)",
+      "Factually accurate (sources verified)",
+      "Meets minimum word count or length requirement",
+      "No spelling or grammar errors",
+      "Complies with legal/regulatory requirements"
+    ],
+    "important": [
+      "Readability score ≥ target (Flesch-Kincaid, etc.)",
+      "Brand voice and tone consistent",
+      "SEO keywords properly integrated (if applicable)",
+      "Target audience appropriately addressed",
+      "Clear structure with headings, paragraphs, lists",
+      "Call-to-action included (if applicable)",
+      "Links and references valid"
+    ],
+    "nice_to_have": [
+      "Engaging storytelling or narrative",
+      "Multimedia elements suggested (images, videos)",
+      "Meta descriptions and titles optimized",
+      "Internal/external linking strategy",
+      "Conversion optimization techniques used"
+    ]
+  },
+  "success_metrics": {
+    "gate_threshold": 0.75,
+    "consensus_threshold": 0.90
+  },
+  "deliverable_requirements": [
+    "Content file created (markdown, doc, html)",
+    "No plagiarism detected",
+    "Factual accuracy verified",
+    "Brand guidelines followed"
+  ]
+}

package/.claude/skills/validation-templates/data.json

@@ -0,0 +1,38 @@
+{
+  "domain": "data-engineering",
+  "validation_criteria": {
+    "critical": [
+      "Data pipeline executes successfully",
+      "Data quality checks pass (completeness, accuracy, consistency)",
+      "Schema validation successful",
+      "No data loss or corruption",
+      "Performance SLAs met"
+    ],
+    "important": [
+      "Data lineage documented",
+      "Error handling and retry logic implemented",
+      "Idempotency ensured (re-runnable without duplicates)",
+      "Monitoring and alerting configured",
+      "Data privacy compliance (PII handling, encryption)",
+      "Partitioning strategy appropriate",
+      "Testing coverage adequate (unit, integration)"
+    ],
+    "nice_to_have": [
+      "Pipeline performance optimized",
+      "Cost efficiency analyzed",
+      "Data catalog entries created",
+      "Data quality metrics tracked over time",
+      "Pipeline documentation comprehensive"
+    ]
+  },
+  "success_metrics": {
+    "gate_threshold": 0.75,
+    "consensus_threshold": 0.90
+  },
+  "deliverable_requirements": [
+    "Pipeline code/configuration",
+    "Data quality validation passed",
+    "Schema documentation",
+    "Performance test results"
+  ]
+}

package/.claude/skills/validation-templates/design.json

@@ -0,0 +1,38 @@
+{
+  "domain": "design",
+  "validation_criteria": {
+    "critical": [
+      "Meets accessibility standards (WCAG 2.1 AA minimum)",
+      "Responsive across target devices/screen sizes",
+      "User flows complete and logical",
+      "Core functionality clearly represented",
+      "Brand guidelines followed"
+    ],
+    "important": [
+      "Design system consistency maintained",
+      "Typography hierarchy clear and effective",
+      "Color contrast ratios meet standards (4.5:1 minimum)",
+      "Touch targets meet minimum size (44x44px)",
+      "Loading states and error states designed",
+      "User testing feedback incorporated",
+      "Component library updated"
+    ],
+    "nice_to_have": [
+      "Microinteractions and animations enhance UX",
+      "Dark mode variant provided",
+      "Print styles considered",
+      "Performance budget respected (image optimization)",
+      "Design tokens documented"
+    ]
+  },
+  "success_metrics": {
+    "gate_threshold": 0.75,
+    "consensus_threshold": 0.90
+  },
+  "deliverable_requirements": [
+    "Design files (Figma, Sketch, Adobe XD)",
+    "Exported assets (if applicable)",
+    "Design specifications document",
+    "Accessibility audit passed"
+  ]
+}

package/.claude/skills/validation-templates/infrastructure.json

@@ -0,0 +1,38 @@
+{
+  "domain": "infrastructure",
+  "validation_criteria": {
+    "critical": [
+      "Infrastructure deploys successfully",
+      "Security posture meets requirements (no exposed secrets, proper IAM)",
+      "Backup and disaster recovery configured",
+      "Monitoring and alerting operational",
+      "Cost within budget"
+    ],
+    "important": [
+      "Infrastructure as Code (IaC) properly structured",
+      "Scalability requirements met",
+      "High availability configured (if required)",
+      "Logging configured and centralized",
+      "Network security groups/firewalls configured",
+      "SSL/TLS certificates valid",
+      "Compliance requirements met (GDPR, HIPAA, etc.)"
+    ],
+    "nice_to_have": [
+      "Auto-scaling configured",
+      "Cost optimization opportunities identified",
+      "Multi-region failover configured",
+      "Performance benchmarks established",
+      "Infrastructure documentation comprehensive"
+    ]
+  },
+  "success_metrics": {
+    "gate_threshold": 0.75,
+    "consensus_threshold": 0.90
+  },
+  "deliverable_requirements": [
+    "IaC files (Terraform, CloudFormation, etc.)",
+    "Deployment successful",
+    "Security scan passed",
+    "Cost estimate provided"
+  ]
+}

package/.claude/skills/validation-templates/research.json

@@ -0,0 +1,38 @@
+{
+  "domain": "research",
+  "validation_criteria": {
+    "critical": [
+      "Research question clearly stated",
+      "Methodology described and appropriate",
+      "Sources cited with proper attribution",
+      "Findings supported by evidence",
+      "No fabricated or falsified data"
+    ],
+    "important": [
+      "Sample size adequate for conclusions",
+      "Statistical methods appropriate and correct",
+      "Limitations acknowledged",
+      "Reproducible methodology",
+      "Peer-reviewed sources prioritized",
+      "Bias assessed and addressed",
+      "Ethical considerations addressed"
+    ],
+    "nice_to_have": [
+      "Novel insights or contributions",
+      "Comprehensive literature review",
+      "Data visualizations clear and effective",
+      "Future research directions suggested",
+      "Practical implications discussed"
+    ]
+  },
+  "success_metrics": {
+    "gate_threshold": 0.75,
+    "consensus_threshold": 0.90
+  },
+  "deliverable_requirements": [
+    "Research report or paper",
+    "Bibliography with cited sources",
+    "Data analysis artifacts (if applicable)",
+    "Methodology documentation"
+  ]
+}

package/.claude/skills/validation-templates/software.json

@@ -0,0 +1,38 @@
+{
+  "domain": "software-development",
+  "validation_criteria": {
+    "critical": [
+      "All tests pass (unit, integration, e2e)",
+      "Security scan shows no CRITICAL or HIGH vulnerabilities",
+      "Build succeeds without errors",
+      "No syntax or runtime errors",
+      "Core functionality works as specified"
+    ],
+    "important": [
+      "Code coverage ≥ 80%",
+      "No linter errors or warnings",
+      "Documentation updated (README, API docs, inline comments)",
+      "Error handling implemented",
+      "Edge cases covered in tests",
+      "No hardcoded secrets or credentials",
+      "Follows project coding standards"
+    ],
+    "nice_to_have": [
+      "Performance benchmarks improved or maintained",
+      "Tech debt reduced (code complexity, duplication)",
+      "Accessibility standards met (WCAG 2.1 AA)",
+      "Logging and monitoring added",
+      "Code is well-refactored and maintainable"
+    ]
+  },
+  "success_metrics": {
+    "gate_threshold": 0.75,
+    "consensus_threshold": 0.90
+  },
+  "deliverable_requirements": [
+    "Source code files created/modified",
+    "Test files with passing tests",
+    "Documentation updated",
+    "No critical security issues"
+  ]
+}

package/.claude/skills/webapp-testing/README.md

@@ -0,0 +1,142 @@
+# Webapp Testing Skill
+
+Visual regression testing with screenshot capture, comparison, and CFN Loop integration.
+
+## Quick Start
+
+### 1. Install Dependencies
+
+```bash
+# Install Playwright browsers
+npx playwright install chromium
+
+# Install Node.js dependencies
+npm install playwright pngjs pixelmatch
+```
+
+### 2. Initialize Storage
+
+```bash
+./.claude/skills/webapp-testing/init-storage.sh
+```
+
+### 3. Run Tests
+
+```bash
+./.claude/skills/webapp-testing/test-webapp-testing.sh
+```
+
+## Usage Examples
+
+### Capture Screenshot
+
+```bash
+./.claude/skills/webapp-testing/capture-screenshot.sh \
+  --project "auth-system" \
+  --component "login-form" \
+  --viewport "1920x1080" \
+  --state "default" \
+  --variant "light-mode" \
+  --url "http://localhost:3000/login" \
+  --task-id "cfn-task-123" \
+  --agent-id "frontend-dev-456"
+```
+
+### Set Baseline
+
+```bash
+./.claude/skills/webapp-testing/set-baseline.sh \
+  --screenshot-key "auth-system/login-form/1920x1080/default/light-mode" \
+  --reason "Initial baseline for login form"
+```
+
+### Compare to Baseline
+
+```bash
+./.claude/skills/webapp-testing/compare-screenshots.sh \
+  --screenshot-key "auth-system/login-form/1920x1080/default/light-mode" \
+  --task-id "cfn-task-123" \
+  --threshold 0.95
+```
+
+## CFN Loop Integration
+
+### Loop 3 (Implementation)
+
+```bash
+# Frontend agent captures screenshots during development
+source ./.claude/skills/webapp-testing/cfn-loop-integration.sh
+
+capture_component_screenshots \
+  "auth-system" \
+  "login-form" \
+  "http://localhost:3000/login" \
+  "$TASK_ID" \
+  "$AGENT_ID" \
+  "1920x1080" "375x667"
+```
+
+### Loop 2 (Validation)
+
+```bash
+# Reviewer compares screenshots and calculates consensus
+source ./.claude/skills/webapp-testing/cfn-loop-integration.sh
+
+compare_all_screenshots "$TASK_ID" 0.95
+CONSENSUS=$(calculate_validation_consensus "$TASK_ID")
+CONFIDENCE=$(echo "$CONSENSUS" | jq -r '.confidence')
+
+echo "Validation confidence: $CONFIDENCE"
+```
+
+### Product Owner Decision
+
+```bash
+# If approved, update baselines
+source ./.claude/skills/webapp-testing/cfn-loop-integration.sh
+
+update_all_baselines \
+  "$TASK_ID" \
+  "Approved visual changes from PR #123" \
+  "product-owner-agent"
+```
+
+## Documentation
+
+- **[SKILL.md](./SKILL.md)** - Complete skill specification
+- **[SCREENSHOT_NAMING_CONVENTION.md](./SCREENSHOT_NAMING_CONVENTION.md)** - Naming and storage strategy
+- **[test-webapp-testing.sh](./test-webapp-testing.sh)** - Validation test suite
+
+## Directory Structure
+
+```
+.screenshots/
+├── baselines/    # Version controlled reference images
+├── current/      # Ephemeral test captures
+├── diffs/        # Comparison outputs
+└── archive/      # Historical captures (90 day TTL)
+```
+
+## Dependencies
+
+- **Playwright** - Browser automation
+- **pngjs** - PNG image parsing
+- **pixelmatch** - Pixel-level diff detection
+- **SQLite** - Metadata storage
+- **Redis** - Agent coordination
+
+## Features
+
+✅ Multi-viewport, multi-state capture
+✅ Pixel-level visual comparison
+✅ Baseline management with audit trail
+✅ CFN Loop integration
+✅ Redis coordination for zero-token waiting
+✅ Hierarchical screenshot naming
+✅ SQLite metadata persistence
+
+## Status
+
+**Version:** 1.0.0
+**Status:** Ready for Testing
+**Last Updated:** 2025-10-24