Weekly recap, May 11, 2026 – May 17, 2026

This week we shipped 47 new step-by-step tutorials for small-business AI, plus the underlying building blocks in 134 npm packages and a new repository. Here’s what you can build right now.

New tutorials

Vercel AI Gateway RAG Knowledge Base for Customer Support

A customer support knowledge base that indexes your help articles and answers questions in real time using Vercel AI Gateway. It ingests content from your CMS, stores vectors in Pinecone, and generates cited answers with conversation history, all while tracking per-query costs.

Read the tutorial → · Download the code (zip)

OpenAI Voice Agent for Appointment Scheduling

Answers after-hours calls, classifies caller intent, and books appointments without human intervention. Uses Twilio, OpenAI Whisper, and Calendly integration, backed by REAA’s agent mesh for orchestration and budget enforcement.

Read the tutorial → · Download the code (zip)

Multi-Agent Observability for Small Business Support

Get a unified dashboard of every AI agent your business runs, tracking latency, cost, and failure patterns. Instruments agents with OpenTelemetry, sends traces to Langfuse, and enables replay of failed interactions.

Read the tutorial → · Download the code (zip)

LangChain Agent Eval Harness for Small Business Reliability

A CLI evaluation harness that lets you systematically test AI agents before they touch customers. Use init, run, and report commands to scaffold configs, execute LangChain-powered evaluations, and export structured reports.

Read the tutorial → · Download the code (zip)

Multi-Agent Handoff Mesh for Small Business Customer Support

Coordinate multiple AI agents with seamless context handoff, budget tracking, and intent routing. An Express server classifies customer messages and routes to specialized agents while preserving conversation state.

Read the tutorial → · Download the code (zip)

Vertex AI Voice Agent for Field Service Dispatch

Automate incoming service calls, classify needs, and route to the right team using a real-time voice agent. Built on Vertex AI, Twilio, and LiveKit, with cost controls and customer history retention.

Read the tutorial → · Download the code (zip)

OpenAI Code Sandbox for SMB Financial Modeling

Securely run AI-generated Python code for financial analysis in isolated E2B sandboxes. An Express server with handoff routing, gateway authentication, and chaos engineering ensures resilient execution.

Read the tutorial → · Download the code (zip)

Anthropic AI Spend Tracker for SMB Agent Workflows

Monitor and control Anthropic API costs across your AI agents with real-time dashboards and per-agent budget enforcement. Auto-downgrades models when limits approach.

Read the tutorial → · Download the code (zip)

Anthropic Document Pipeline for SMB Accounting

Automated invoice processing that extracts, validates, and posts data from PDFs into QuickBooks. Uses Claude, Redis-backed queues, and per-invoice cost tracking.

Read the tutorial → · Download the code (zip)

Vertex AI Budget Guardrails for Multi-Agent Systems

Keep your AI costs predictable with per-agent budget limits and automatic model routing on Vertex AI. An Express server enforces spend controls and provides real-time metrics.

Read the tutorial → · Download the code (zip)

AWS Bedrock AI Incident Runbook for DevOps-light SMBs

Automated incident response and rollback for AI agents, with plain-language triage via AWS Bedrock. Monitors services, classifies failures, triggers runbooks, and sends Slack alerts.

Read the tutorial → · Download the code (zip)

OpenRouter Lead Intake Agent for Small Business

Never lose a lead again. An AI intake agent classifies, enriches, and hands off leads to your sales team 24/7, with SMS alerts for high-priority submissions.

Read the tutorial → · Download the code (zip)

AWS Bedrock MCP Server for Small Business DevOps Runbooks

An MCP server that exposes AWS Bedrock‑powered runbook automation to any AI assistant. Self‑healing infrastructure tools without a dedicated DevOps team.

Read the tutorial → · Download the code (zip)

Azure AI Agent Auth Gateway for Multi-Tenant SMB SaaS

A drop-in authentication proxy that forces every AI agent tool call to carry a valid Azure AD token, securing multi-tenant SMB apps by default.

Read the tutorial → · Download the code (zip)

OpenAI Knowledge Agent for SMB Employee Onboarding

A persistent AI memory system that ingests onboarding docs, learns company norms, and answers new hires’ questions. Uses OpenAI embeddings and pgvector to search policy documents with citations.

Read the tutorial → · Download the code (zip)

Azure AI Multi-Agent Handoff for E-commerce Customer Support

A routing mesh that classifies customer intents and hands off conversations between specialist agents, all hosted on Azure AI. Scales support without hiring.

Read the tutorial → · Download the code (zip)

OpenRouter Security Guardrails for SMB API Protection

A lightweight API gateway that screens every AI prompt and response for PII, injection attempts, and unsafe content, using OpenRouter’s unified model access.

Read the tutorial → · Download the code (zip)

Vercel AI Gateway Reliability Suite for SMB AI Operations

A self‑serve reliability dashboard that monitors, replays, and self‑heals AI workflows running through Vercel AI Gateway. Records every LLM interaction for replay.

Read the tutorial → · Download the code (zip)

LangChain Observability for SMB AI Workflow Monitoring

Plug‑and‑play tracing and cost observability for LangChain‑based pipelines, built on REAA’s OpenTelemetry stack. Exports traces to Langfuse for real-time monitoring.

Read the tutorial → · Download the code (zip)

Anthropic Eval Harness for Agent Quality Assurance

Continuous regression testing and safety scoring for Anthropic‑powered agents, with automated quality gates before any customer‑facing deployment.

Read the tutorial → · Download the code (zip)

AWS Bedrock Lead Intake for Small Business Growth

Phone calls and web forms become structured, qualified leads—automatically routed to your CRM with AI voice analysis and document extraction.

Read the tutorial → · Download the code (zip)

xAI Grok PII Detection for SMB Customer Communication

Protect customer data by scanning Grok-powered messages for PII and offensive content before they reach end users, via an Express middleware proxy.

Read the tutorial → · Download the code (zip)

Mistral AI Invoice Extraction for SMB Accounting

Automatically extract vendor, amount, and line items from invoices using Mistral Large, with cost-aware processing and human review for low-confidence results.

Read the tutorial → · Download the code (zip)

Cohere RAG Legal Research for SMB Law Firms

Instant case law Q&A with citations, powered by Cohere embeddings and retrieval. Indexes documents into Qdrant and answers legal questions with budget enforcement.

Read the tutorial → · Download the code (zip)

Anthropic Code Sandbox for SMB Data Analysis

Turn plain-English questions into safe, budget-controlled data analyses in a secure E2B sandbox. Upload CSVs and get answers from Claude, with per-query spend limits.

Read the tutorial → · Download the code (zip)

Mistral AI Invoice Reconciliation for Stripe

Automatically extracts structured data from PDF invoices and reconciles them against Stripe transactions, flagging discrepancies for your finance team.

Read the tutorial → · Download the code (zip)

AWS Bedrock Multi-Agent Handoff for Zendesk Ticket Triage

Route Zendesk support tickets to specialized AI agents based on topic, with seamless context handoff and budget controls per ticket.

Read the tutorial → · Download the code (zip)

Google Gemini Reliability Suite for SMB AI Operations

Resilient Gemini-powered agents with built-in circuit breakers, idempotency, incident runbooks, and budget controls. Prevents duplicate operations and handles tool failures gracefully.

Read the tutorial → · Download the code (zip)

OpenRouter Agent Mesh for SMB Workflow Automation

Coordinate multiple specialist AI agents across teams using a single low-cost routing layer with observability and spend controls. Dispatches tasks to the right model via OpenRouter.

Read the tutorial → · Download the code (zip)

Azure AI Document Pipeline for QuickBooks Online Invoice Processing

Automatically extract, validate, and post vendor invoice data from PDFs or images directly into QuickBooks Online using Azure OpenAI, with PII redaction.

Read the tutorial → · Download the code (zip)

Anthropic Voice Agent for Calendly Appointment Booking

A voice assistant that uses Calendly’s API to schedule appointments, handling natural language availability queries and booking conflicts with Anthropic models.

Read the tutorial → · Download the code (zip)

Mistral AI Multi-Agent Handoff for Linear Support Triage

Route, hand off, and escalate customer support issues in Linear with a Mistral-powered multi-agent system that preserves context across specialist agents.

Read the tutorial → · Download the code (zip)

AWS Bedrock Cost Control for Serverless AI Applications

A budget‑aware middleware that routes, caps, and tracks every AWS Bedrock LLM call, preventing surprise bills for serverless AI features.

Read the tutorial → · Download the code (zip)

Anthropic Security Guardrails for SMB AI Chatbots

Pluggable Express middleware that scrubs PII, blocks prompt injections, and enforces content policies on Anthropic-powered chatbots.

Read the tutorial → · Download the code (zip)

Ollama RAG Knowledge Base for Retail Inventory

Self-hosted knowledge base that lets retail staff query product inventory docs using natural language, fully local with Ollama, LanceDB, and fastembed.

Read the tutorial → · Download the code (zip)

vLLM Security Guardrails for SMB API Gateways

A drop-in API proxy that adds PII redaction, prompt injection defense, and content safety checks to any vLLM endpoint. Includes an admin dashboard.

Read the tutorial → · Download the code (zip)

Vertex AI Voice Agent for Cal.com Appointment Scheduling

Let customers book appointments on Cal.com over the phone with a voice agent that understands natural language and confirms bookings in real time.

Read the tutorial → · Download the code (zip)

xAI Grok Knowledge Agent for SMB HR Policy Q&A

Empower employees to self‑serve PTO, benefits, and handbook questions with a Grok‑backed agent that remembers policy contexts across sessions.

Read the tutorial → · Download the code (zip)

Anthropic RAG Knowledge Base for Construction Project Specs

Let field crews instantly query building specifications, codes, and submittals via a Claude‑powered hybrid retrieval pipeline with BM25 and vector search.

Read the tutorial → · Download the code (zip)

Perplexity Market Research Document Pipeline for SMBs

Transform scattered market reports into a searchable research assistant powered by your documents and live market data. Ingests PDFs, DOCX, and XLSX into Qdrant.

Read the tutorial → · Download the code (zip)

OpenAI Multi-Agent Handoff for E-commerce Order Tracking

Automate order‑status, shipping, and returns inquiries by routing customers among specialized AI agents without losing conversation context.

Read the tutorial → · Download the code (zip)

Cohere Code Sandbox for SMB Data Workflows

Secure, budget-controlled code execution environment for SMB analysts, using Cohere’s models and E2B sandbox with spend enforcement.

Read the tutorial → · Download the code (zip)

Google Gemini MCP Server for SMB Business Intelligence

A turnkey MCP server that lets SMB teams use natural language to query their business data from HubSpot, Stripe, and Google Workspace via a Gemini-powered interface.

Read the tutorial → · Download the code (zip)

Cohere Lead Intake Agent for HubSpot SMB Sales

An AI intake system that processes documents, emails, and chat messages to capture leads and auto-populate HubSpot CRM with categorized data.

Read the tutorial → · Download the code (zip)

Azure AI Phone Agent for Cal.com Appointment Scheduling

An AI voice agent that answers calls, converses with customers, and books appointments directly into Cal.com, using Azure OpenAI and REAA’s orchestration toolkit.

Read the tutorial → · Download the code (zip)

Perplexity Knowledge Agent for Notion SMB FAQ

A conversational AI that answers employee and customer questions by searching Notion workspaces, using Perplexity’s search and hybrid RAG for cited responses.

Read the tutorial → · Download the code (zip)

Agnostic AI Budget Control for SMB Agent Workflows

Real-time spend enforcement and cost-aware routing across any LLM provider for small business AI agents. Caps per-session and monthly costs with dynamic model selection.

Read the tutorial → · Download the code (zip)

New repository

reaatech

A modular TypeScript toolkit for building and deploying AI-driven business systems — voice agents, RAG pipelines, automated workflows — with no vendor lock-in. Every package is MIT-licensed and self-hosted. Browse the code · Catalog page

Building blocks shipped

Agent Runbook Generator

Automated runbook generation and incident response for cloud services. New packages include core types and schemas (@reaatech/agent-runbook), OpenTelemetry observability (agent-runbook-observability), an analysis agent for LLMs (agent-runbook-agent), and a CLI with MCP server mode (agent-runbook-cli). Browse all →

AGENTS.md Kit

Lint, validate, scaffold, and manage AGENTS.md and SKILL.md files for AI agents. Shipped with a parser, linter, validator, reporter, and MCP server. Browse all →

Circuit Breaker Agents

Circuit breaker state machine for agent-to-agent reliability, with pluggable persistence adapters (Redis, Firestore, DynamoDB). Browse all →

Classifier Evals

Evaluation framework for classification models: dataset loading, metrics, quality gates, LLM-as-judge, and CLI. Browse all →

Confidence Router

Decision engine for routing based on prediction confidence, with language support and threshold optimization. Browse all →

Context Window Planner

Manage LLM context windows with token-aware packing strategies and a CLI. Browse all →

Guardrail Chain

Orchestrate sequences of LLM guardrails with budget management and circuit breaking. Includes 13 built-in guardrails for PII, prompt injection, and content safety. Browse all →

Hybrid RAG (Qdrant)

Full hybrid RAG pipeline with ingestion, embedding, retrieval, evaluation, and observability, backed by Qdrant. Browse all →

Idempotency Middleware

Prevent duplicate execution of side-effecting operations with pluggable storage adapters (Redis, DynamoDB, Firestore) and Express/Koa middleware. Browse all →

LLM Cache

Multi-stage caching for LLM responses with semantic similarity, exact-match lookups, and cost tracking. Adapters for Redis, DynamoDB, Qdrant. Browse all →

LLM Cost Telemetry

Provider-agnostic cost calculation, budgeting, and OpenTelemetry export for LLM usage, with wrappers for OpenAI, Anthropic, and Gemini. Browse all →

LLM Judge Toolkit

Evaluate LLM responses with pluggable providers, consensus aggregation, calibration, and bias detection. Browse all →

LLM Router

Dynamic model routing with budget control, fallback chains, and A/B testing, plus an MCP server for agent integration. Browse all →

MCP Catalog Shared

Shared types and Zod schemas for MCP catalog services. Browse →

MCP Contract Kit

Validate MCP server compliance, security, and performance with contract testing, reporting, and a CLI. Browse all →

MCP Gateway

A full-featured MCP gateway with auth, rate limiting, schema validation, caching, fanout, and observability. Browse all →

MCP Load Test

Load testing for MCP servers with configurable profiles, metrics, and breaking point analysis. Browse all →

MCP Schema Evolution

Compare MCP tool schemas and detect breaking changes across versions, with CLI and CI tools. Browse all →

MCP Server Doctor

Diagnostic checks for MCP servers with grading, latency profiling, and schema validation. Browse all →

MCP Server Starter TS

Framework for building MCP servers on Express 5 with auth, observability, tool registration, and transport handling. Browse all →

Browse the full catalog at reaatech.com/products.