jeremylongshore/claude-code-plugins-plus-skills v1.3.1

v1.3.1 - Google AI Plugins Major Update

on GitHub

[1.3.1] - 2025-11-15

🎉 Highlights

MASSIVE UPDATE - Google AI plugins completely rewritten (3,483 lines), 8 new plugins added, automated asset generation system, comprehensive audit infrastructure, and CI/CD improvements.

Breaking Record: Largest single release in repository history with 24 commits spanning Nov 9-15, 2025.

📊 Release Statistics

Scope of Changes:

• 24 commits across 7 days
• 8 new plugins added (Google Cloud, Vertex AI, Firebase ecosystem)
• 5 plugins updated (Google AI suite with 3,483 lines added)
• 231 plugins receive automated asset generation
• 261+ assets generated via Gemini AI (target: ~1,500)
• 100% README coverage achieved (253/253 plugins)
• 9 catalog sync commits for marketplace validation
• 2 CI/CD improvements (workflow validation, plugin fixes)

Lines Changed:

• +3,483 lines documentation in Google AI plugins
• +297 lines jeremy-firebase README
• +800 lines asset generation system (Python, SQLite)
• +2,500 lines audit reports and analysis

Quality Improvements:

• 92% of plugins identified with incomplete assets (audit)
• 100% plugin coverage in CI/CD validation (was ~40%)
• Zero asset generation failures with rate limiting
• 2025 schema compliance across all new plugins

🏆 Major Achievement

Agent Engine vs Cloud Run Clarity - Eliminated all confusion about deployment targets across the Google AI plugin suite.

🔧 Plugins Updated (5/5 Complete)

1. jeremy-vertex-engine (v1.0.0 → v1.0.1)

• Added: 403 lines (+113% growth)
• New Sections:
  • Observability & Monitoring (174 lines) - 2025 Agent Engine dashboard, Cloud Trace, custom metrics
  • Storage Integration (239 lines) - BigQuery connector, Cloud Storage, export patterns
  • Prerequisites & Dependencies (144 lines) - Complete setup guide
• Features: 15+ production code examples, 3 SQL analytics queries, 5 alert policy patterns

2. jeremy-adk-orchestrator (v1.0.0 → v1.0.1) - CRITICAL FIX

• Added: 695 lines (README created from scratch - was completely missing!)
• New Content:
  • Complete A2A Protocol documentation (AgentCard discovery, task submission, status polling)
  • Multi-agent orchestration patterns (supervisory architecture)
  • Observability integration (Cloud Trace, Logging, custom metrics)
  • BigQuery storage export for orchestration analytics
• Features: 40+ production Python code examples

3. jeremy-vertex-validator (v1.0.0 → v1.0.1)

• Added: 662 lines (+2006% growth from 33 lines)
• New Sections:
  • 5-category validation system (470 lines) - Security, Monitoring, Performance, Compliance, Best Practices
  • Production readiness report format (120 lines) - Weighted scoring with example 87-line report
  • Validation code examples (150+ lines) - Python validation functions
• Features: Security checks (IAM, VPC-SC, Model Armor), weighted scoring (0-100%)

4. jeremy-adk-terraform (v1.0.0 → v1.0.1)

• Added: 827 lines (was 35 lines with wrong content)
• Infrastructure Modules:
  • Core Agent Engine resource (65 lines HCL)
  • IAM configuration with least privilege (40 lines)
  • VPC Service Controls (25 lines)
  • Observability infrastructure (155 lines) - Dashboards, alerts, 2025 features
  • BigQuery analytics (105 lines) - Log sink, partitioned tables, sink IAM
  • Cloud Storage artifacts (60 lines) - Lifecycle policies, versioning
• Features: Complete terraform modules for Agent Engine deployment

5. jeremy-vertex-terraform (v1.0.0 → v1.0.1)

• Added: 896 lines (was 35 lines with wrong content)
• Infrastructure Modules:
  • Gemini API endpoints (67 lines HCL)
  • Vector Search infrastructure (95 lines) - ScaNN-based similarity search for RAG
  • Custom model deployment (100 lines) - GPU serving endpoints
  • Vertex AI Pipelines (80 lines) - Kubeflow integration
  • Feature Store (50 lines) - ML feature management
  • Batch prediction (40 lines) - Large-scale inference
  • Monitoring dashboards (145 lines) - 4 widgets, 2 alert policies
• Features: Model Garden, Gemini, Vector Search, Pipelines, Feature Store

📊 Total Documentation Added

• 3,483 lines of production-grade documentation
• 60+ working code examples (Python, HCL, SQL, Bash)
• 7 complete Terraform modules for Agent Engine
• 7 complete Terraform modules for Vertex AI services
• 3 BigQuery analytics queries with partitioning
• 10+ alert policy configurations

🎯 Key Improvements Across All Plugins

✅ Deployment Target Clarity

• Prominent 🎯 headers ("VERTEX AI AGENT ENGINE ONLY" vs "MODEL GARDEN & AI INFRASTRUCTURE")
• Clear ✅ THIS PLUGIN IS FOR / ❌ THIS PLUGIN IS NOT FOR sections
• Eliminated Cloud Run vs Agent Engine confusion

✅ Complete Prerequisites & Dependencies

• Google Cloud API enablement commands
• Authentication setup (ADC and service accounts)
• IAM permissions with specific role names
• Python packages with minimum version requirements
• gcloud CLI installation and verification
• Terraform provider configuration

✅ 2025 Observability Features (NEW)

• Agent Engine Observability Dashboard access
• Cloud Trace integration with OpenTelemetry
• Cloud Logging structured queries
• Custom metrics and alerting policies
• Token usage and cost tracking

✅ 2025 Storage Integration (NEW)

• BigQuery connector for agent logs
• Cloud Storage integration for artifacts
• Data export patterns (real-time, daily, compliance)
• Analytics SQL queries
• Scheduled reports

🆕 New Plugins Added (8 Total)

This release includes 8 brand new plugins spanning Google Cloud operations, Vertex AI tooling, and Firebase development.

1. jeremy-firestore (v1.0.0) - commit 5323f25

Comprehensive Firestore plugin with A2A/MCP/Cloud Run support

• Agent: jeremy-firestore:firestore-operations-expert - Production Firestore data modeling, security rules, and performance optimization
• Features:
  • Advanced data modeling patterns (denormalization, subcollections, composite indexes)
  • Security rules validation and audit
  • Batch operations with transaction management
  • Real-time listener optimization
  • Cloud Functions integration for Firestore triggers
  • Vertex AI Gemini integration for AI-powered data analysis
  • A2A protocol support for agent orchestration
  • MCP tools for programmatic Firestore access
• Documentation: 297-line comprehensive README with production examples
• Use Cases: Firebase web/mobile apps, real-time dashboards, serverless backends

2. jeremy-github-actions-gcp (v1.0.0) - commit e8d456f

GitHub Actions with Workload Identity Federation enforcement and Vertex AI validation

• Agent: jeremy-github-actions-gcp:gh-actions-gcp-expert - CI/CD pipeline automation for GCP deployments
• Skill: jeremy-github-actions-gcp:gh-actions-validator - Validates workflow files for WIF compliance
• Features:
  • Workload Identity Federation (WIF) enforcement - NO service account keys allowed
  • Automated Vertex AI Agent Engine deployments
  • Cloud Run deployment pipelines
  • Secret management with Google Secret Manager
  • Comprehensive GitHub Actions best practices
  • Security scanning integration
  • Multi-environment deployment strategies
• Security: Zero static credentials, OIDC-based authentication only
• Use Cases: Secure GCP deployments, agent CI/CD, multi-stage pipelines

3. jeremy-gcp-starter-examples (v1.0.0) - commit fe8e09b

Official Google Cloud starter kits, ADK samples, Genkit templates, and Vertex AI examples

• Agent: jeremy-gcp-starter-examples:gcp-starter-kit-expert - Official GCP code samples and best practices
• Features:
  • Agent Starter Pack (official Google agent templates)
  • Firebase Genkit flow examples (Node.js, Python, Go)
  • Vertex AI ADK sample applications
  • Cloud Run quickstart templates
  • BigQuery data pipeline examples
  • Terraform infrastructure samples
  • Production-ready configuration examples
• Content: Curated from official Google Cloud repositories
• Use Cases: Quick project bootstrapping, reference implementations, learning GCP patterns

4. jeremy-vertex-engine (v1.0.0) - commit 8125c14

Comprehensive Vertex AI Agent Engine inspector and operations toolkit

• Agent: jeremy-vertex-engine:vertex-engine-inspector - Validates Agent Engine deployments and runtime health
• Features:
  • Agent Engine deployment validation
  • Runtime configuration inspection (Code Execution Sandbox, Memory Bank)
  • A2A protocol compliance checking
  • Production readiness assessment
  • Health monitoring and diagnostics
  • Performance metrics analysis
  • Security posture validation (IAM, VPC-SC, Model Armor)
• Validation Categories: Runtime config, A2A compliance, sandbox settings, memory bank, security, observability
• Use Cases: Pre-deployment checks, production monitoring, compliance audits

5. overnight-dev (v1.0.0) - commit 78cd711

Overnight development automation for unattended project work

• Agent: overnight-dev:overnight-automation - Automates long-running development tasks
• Features:
  • Unattended code generation and refactoring
  • Automated test suite execution
  • Documentation generation during off-hours
  • Database migration execution
  • Build and deployment automation
  • Error handling and rollback
  • Morning summary reports
• Safety: Comprehensive pre-flight checks, automatic backups, rollback on failure
• Use Cases: Large refactoring projects, batch processing, overnight builds

6. jeremy-vertex-search (v1.0.0) - commit 748b3e7

Vertex AI Search and RAG (Retrieval-Augmented Generation) implementation toolkit

• Agent: jeremy-vertex-search:vertex-search-expert - Enterprise search and RAG patterns
• Features:
  • Vertex AI Search configuration and indexing
  • Vector Search with ScaNN similarity search
  • RAG pipeline implementation (ingestion, chunking, retrieval, generation)
  • Document parsing and preprocessing
  • Query optimization and reranking
  • Grounding with Google Search
  • Multi-modal search (text, images, video)
• Infrastructure: BigQuery for metadata, Cloud Storage for documents, Vector Search for embeddings
• Use Cases: Enterprise search, chatbots with context, document Q&A, semantic search

7. jeremy-vertex-observability (v1.0.0) - commit 748b3e7

2025 Vertex AI observability and telemetry features

• Agent: jeremy-vertex-observability:observability-expert - Monitoring, logging, and tracing for Vertex AI
• Features:
  • Agent Engine Observability Dashboard (2025 feature)
  • Cloud Trace integration with OpenTelemetry
  • Cloud Logging structured query templates
  • Custom metrics and alerting policies
  • Token usage and cost tracking
  • Performance profiling
  • Real-time agent behavior monitoring
• Telemetry: Request tracing, latency analysis, error rate tracking, resource utilization
• Use Cases: Production monitoring, performance optimization, cost management, debugging

8. jeremy-vertex-storage (v1.0.0) - commit 748b3e7

Vertex AI storage integration - BigQuery connector and Cloud Storage patterns

• Agent: jeremy-vertex-storage:storage-integration-expert - Data storage patterns for Vertex AI
• Features:
  • BigQuery connector for agent logs
  • Cloud Storage integration for artifacts
  • Data export patterns (real-time, daily, compliance)
  • Analytics SQL queries for agent behavior
  • Scheduled data exports
  • Lifecycle policies and retention management
  • Data partitioning and clustering strategies
• Integration: Vertex AI → BigQuery, Cloud Storage for model artifacts, backup strategies
• Use Cases: Agent log analytics, compliance reporting, data warehousing, audit trails

🤖 Asset Generation System (NEW)

Major Innovation: Automated context-aware asset file generation using Vertex AI Gemini.

Script: scripts/generate-missing-assets.py

• AI Model: Gemini 1.5 Flash (gemini-1.5-flash-002) for stable, high-quota generation
• Context-Aware: Reads each plugin's README.md and plugin.json for accurate content generation
• Multi-Format Support: JSON, YAML, Markdown, HTML, Python, Shell scripts
• Rate Limiting: 2-second delay between API calls with exponential backoff (5s → 15s → 45s)
• Progress Tracking: SQLite database (backups/asset_generation.db) tracks all generation attempts
• Retry Logic: 3 attempts per file with intelligent backoff on quota errors
• Safety: Validates generated content before writing to disk

Performance:

• Generated 261+ assets successfully (as of Nov 15)
• Target: ~1,500 total assets across 231 plugins
• Zero failures with rate limiting enabled
• Average file size: 2,500 bytes (context-rich, production-ready)

Example Output (web-to-github-issue plugin):

assets/
├── issue_template.md (2,662 bytes) - GitHub issue format with web research context
├── config_template.json (2,514 bytes) - API settings, rate limiting, security configs
└── example_search_results.json (2,520 bytes) - Web search result format

Commits: e72aed1, bd57367

📋 Asset Audit Infrastructure (NEW)

Comprehensive Plugin Audit System identifying incomplete asset directories across the entire marketplace.

Audit Report: asset-audit-report-20251115.txt

• Scope: 252 plugins analyzed
• Findings: 231 plugins (92%) with incomplete asset directories
• Impact Analysis: ASSET_AUDIT_RESPONSE.md with root cause and recommendations

Key Statistics:

• 76 plugins missing SKILL.md files
• 230 plugins with incomplete asset directories
• Most common missing files:
  • report_template.html (19 plugins)
  • example_dataset.csv (8 plugins)
  • config_template.json (5 plugins)
  • api_template.yaml (4 plugins)

Root Cause: Automated plugin generation created assets/README.md with checklists but never generated the actual files.

Resolution Strategy:

• Short-term: Gemini-powered automated generation (in progress)
• Medium-term: Template library for common asset types
• Long-term: Enhanced plugin validation in CI/CD

Commit: 547f2b6

🔧 CI/CD Improvements

GitHub Workflow Validation Enhancement - commit 60ed152

Problem: Workflow only validated plugins/community/* and plugins/examples/*, missing 200+ plugins in other categories.

Solution: Dynamic plugin discovery using find command

# Before: Hardcoded paths
for plugin in plugins/community/*/ plugins/examples/*/; do

# After: Dynamic discovery
find plugins/ -name "plugin.json" -path "*/.claude-plugin/plugin.json" | while read plugin_json; do
  plugin=$(dirname $(dirname "$plugin_json"))

Impact: Now validates all 253 plugins regardless of directory structure.

Security Checks Added:

• Hardcoded secrets detection
• AWS keys detection (blocks CI)
• Private key detection (blocks CI)
• Dangerous command patterns (rm -rf /)
• Command injection risks (eval())
• Suspicious URLs (non-HTTPS, URL shorteners)
• MCP plugin dependency audit (npm audit)

devops-automation-pack Plugin Fix - commit 4ff8059

Problem: Invalid manifest with non-existent path references

"commands": "./plugins/*/commands",  // Invalid wildcard paths
"agents": "./plugins/*/agents"

Solution: Removed invalid fields, kept only valid component declarations

Impact: Plugin now passes marketplace validation

📚 README Coverage Achievement

100% README Coverage - All 253 plugins now have comprehensive documentation.

Jeremy-Firebase README Created - commit 57f3523

• Size: 297 lines of comprehensive documentation
• Content:
  • Detailed feature overview
  • Vertex AI Gemini integration examples
  • Security rules patterns
  • Production deployment guide
  • A2A protocol usage
  • MCP tools reference
  • Code examples for all major features

Cleanup: Removed duplicate empty fairdb-operations-kit/ directory (real plugin exists in plugins/devops/)

Impact: Asset generation can now proceed for all plugins (README provides essential context)

🔄 Marketplace Catalog Updates

Multiple catalog synchronization and validation improvements across 9 commits.

Commits: 071d2ee, 438f8e0, b3fd9cc, 7278c1b, b2f5efa, 0e6091a, 904f360, f6c190a, 29c5a63

Improvements:

• Synchronized marketplace.extended.json with marketplace.json
• Updated plugin metadata for new releases
• Fixed category assignments
• Validated all plugin references
• Updated plugin counts and statistics
• Ensured schema compliance

Quality Assurance: All changes validated by CI/CD pipeline before merge

✅ Skills Updated

• jeremy-adk-orchestrator/adk-deployment-specialist: v1.0.0 → v1.0.1 (A2A protocol focus)
• jeremy-vertex-validator/validator-expert: v1.0.0 → v1.0.1 (2025 features)
• jeremy-firestore/firestore-operations-expert: v1.0.0 (NEW)
• jeremy-github-actions-gcp/gh-actions-gcp-expert: v1.0.0 (NEW)
• jeremy-gcp-starter-examples/gcp-starter-kit-expert: v1.0.0 (NEW)
• jeremy-vertex-engine/vertex-engine-inspector: v1.0.0 (NEW)
• overnight-dev/overnight-automation: v1.0.0 (NEW)
• jeremy-vertex-search/vertex-search-expert: v1.0.0 (NEW)
• jeremy-vertex-observability/observability-expert: v1.0.0 (NEW)
• jeremy-vertex-storage/storage-integration-expert: v1.0.0 (NEW)

🔑 Critical Distinctions Documented

jeremy-adk-terraform vs jeremy-vertex-terraform:

Both plugins now have:

• Clear scope definitions
• Non-overlapping use cases
• Cross-references to each other
• Complete Terraform examples

📝 Supporting Documentation

New Release Notes: plugins/community/jeremy-firebase/000-usermanuals/010-plugin-updates-release-notes.md

• 30KB comprehensive release documentation
• Breaking changes (none - documentation only)
• Migration guide
• Testing checklist
• Communication plan
• Success criteria

🛠️ Dependencies Added/Updated

Python Packages:

google-cloud-aiplatform[agent_engines]>=1.120.0
google-adk>=1.15.1
google-cloud-logging>=3.10.0
google-cloud-monitoring>=2.21.0
google-cloud-trace>=1.13.0
a2a-sdk>=0.3.4
google-cloud-security-center>=1.28.0 (validator)
pylint>=3.0.0, flake8>=7.0.0, mypy>=1.8.0 (validator)

Terraform Requirements:

terraform >= 1.5.0
google provider >= 5.0
google-beta provider >= 5.0

Google Cloud APIs:

• aiplatform.googleapis.com
• discoveryengine.googleapis.com
• logging.googleapis.com
• monitoring.googleapis.com
• cloudtrace.googleapis.com
• bigquery.googleapis.com
• storage.googleapis.com

🚀 Production Features

Agent Engine Terraform:

• Code Execution Sandbox (1-14 days TTL validation)
• Memory Bank (retention policies, max memories)
• VPC Service Controls perimeter
• IAM least privilege
• Auto-scaling configuration
• Model Armor (prompt injection protection)
• CMEK encryption

Vertex AI Terraform:

• Model Garden deployments (Gemini, PaLM, Claude, Llama)
• Vector Search with ScaNN algorithm
• Feature Store for ML features
• Batch prediction jobs with GPUs
• ML Pipelines with Kubeflow
• Traffic splitting for blue/green deployments

📈 Observability Features (2025)

Cloud Monitoring Dashboards:

• Request volume
• Error rates
• Latency distribution (p50, p90, p95, p99)
• Token usage and cost estimation
• Memory Bank cache hit rate
• Code Execution Sandbox stats
• Replica utilization

Alert Policies:

• High error rate (>5% for 5 minutes)
• High latency (p95 > 10s)
• Token budget exceeded
• Memory Bank cache degradation
• Code Execution timeout rate

BigQuery Analytics:

• Agent query volume and latency trends
• Memory Bank cache performance
• Token usage and cost analysis
• Multi-agent workflow patterns
• Component-level metrics (AGENT_QUERIES, MEMORY_BANK_OPERATIONS, CODE_EXECUTION_EVENTS)

🎓 Use Cases Documented

Agent Engine:

• Pre-production validation
• Post-deployment verification
• Security audits
• Multi-agent orchestration
• A2A protocol communication

Vertex AI:

• Gemini API deployment
• Vector search for RAG
• Custom model serving
• Batch predictions
• Feature Store setup

👥 Contributors

Thanks to @jeremylongshore for the comprehensive Google AI plugin suite overhaul!

📦 Installation

# Install all 13 plugins (5 updated + 8 new)
/plugin install jeremy-vertex-engine@claude-code-plugins-plus
/plugin install jeremy-adk-orchestrator@claude-code-plugins-plus
/plugin install jeremy-vertex-validator@claude-code-plugins-plus
/plugin install jeremy-adk-terraform@claude-code-plugins-plus
/plugin install jeremy-vertex-terraform@claude-code-plugins-plus
/plugin install jeremy-firestore@claude-code-plugins-plus
/plugin install jeremy-github-actions-gcp@claude-code-plugins-plus
/plugin install jeremy-gcp-starter-examples@claude-code-plugins-plus
/plugin install overnight-dev@claude-code-plugins-plus

🔗 Complete GitHub Commit History (24 commits)

Nov 15, 2025 - Asset Generation & GitHub Release:

• bd57367: feat: enhance asset generation with rate limiting and retry logic
• e72aed1: feat: add comprehensive asset generation system with Gemini AI
• 57f3523: docs: add comprehensive jeremy-firebase README and fix duplicate fairdb directory

Nov 13, 2025 - CI/CD & Marketplace Fixes:

• 60ed152: fix: update GitHub workflow to validate all plugins dynamically
• 4ff8059: fix: remove invalid manifest fields from devops-automation-pack

Nov 13, 2025 - Google AI Plugins Major Update:

• 7a2c06e: docs: update CHANGELOG for Google AI plugins v1.0.1
• 1f76301: feat: complete jeremy-vertex-terraform with comprehensive Vertex AI infrastructure
• 528744a: feat: complete jeremy-adk-terraform with Agent Engine Terraform modules
• 237400f: feat: comprehensive update to jeremy-vertex-engine, jeremy-adk-orchestrator, jeremy-vertex-validator

Nov 13, 2025 - Marketplace Catalog Synchronization:

• 29c5a63: chore: sync marketplace catalogs
• f6c190a: chore: sync marketplace catalogs
• 904f360: chore: sync marketplace catalogs
• 0e6091a: chore: sync marketplace catalogs
• b2f5efa: chore: sync marketplace catalogs
• 7278c1b: chore: sync marketplace catalogs
• b3fd9cc: chore: sync marketplace catalogs
• 438f8e0: chore: sync marketplace catalogs
• 071d2ee: chore: sync marketplace catalogs

Nov 10, 2025 - New Plugins Added:

• 5323f25: feat(jeremy-firestore): add killer Firestore plugin with A2A/MCP/Cloud Run support
• 78cd711: fix(overnight-dev): remove invalid 'featured' field from plugin manifest
• e8d456f: feat: add jeremy-github-actions-gcp plugin with WIF enforcement and Vertex AI validation
• fe8e09b: feat: add jeremy-gcp-starter-examples plugin with official Google Cloud code samples
• 8125c14: feat: add jeremy-vertex-engine - comprehensive Agent Engine inspector

Nov 9, 2025 - Asset Audit Infrastructure:

• 547f2b6: feat: add comprehensive plugin asset audit system