Weekly recap, June 8, 2026 – June 14, 2026

This week we published 53 new step-by-step tutorials for small-business AI — from extracting invoice data into QuickBooks to automating no-show enforcement for fitness studios. Every one includes a downloadable zip, a test suite, and no contact-us-for-a-quote required.

New tutorials

Databricks MCP Server for SMB Data Analytics

Wrap Databricks SQL queries as secure, multi-tenant tools for AI agents. The server enforces per-tenant isolation, rate limits, and auth — letting SMBs give agents safe access to their BI data without building the plumbing from scratch.

Read the tutorial → · Download the code (zip)

Built with @reaatech/mcp-server-core, @reaatech/mcp-server-engine, @reaatech/mcp-server-auth, @reaatech/mcp-server-observability, @reaatech/multi-tenant-mcp-middleware, @reaatech/multi-tenant-mcp-tenant-resolver, @reaatech/multi-tenant-mcp-rate-limiter on Databricks · 85 tests / 98.75% coverage.

xAI Grok Invoice Extraction for QuickBooks SMBs

Extract structured invoice data from PDFs using xAI Grok and push it straight into QuickBooks. The built-in repair layer fixes malformed JSON, cost telemetry tracks spend per invoice, and budget controls prevent runaway bills.

Read the tutorial → · Download the code (zip)

Built with @reaatech/structured-repair-core, @reaatech/llm-cost-telemetry, @reaatech/llm-cost-telemetry-calculator, @reaatech/agent-budget-engine, @reaatech/agent-budget-spend-tracker, @reaatech/idempotency-middleware on xAI Grok · 83 tests / 98.88% coverage.

AI Runbook Automation for Agent Failure Recovery

Automatically trigger health checks and runbooks when AI agents fail. Circuit breakers isolate broken dependencies, idempotent retries make recovery safe, and the whole system works across any LLM provider.

Read the tutorial → · Download the code (zip)

Built with @reaatech/agent-runbook, @reaatech/agent-runbook-incident, @reaatech/agent-runbook-health-checks, @reaatech/circuit-breaker-core, @reaatech/circuit-breaker-agents, @reaatech/idempotency-middleware · 115 tests / 94.59% coverage.

Google Gemini Lead Intake for Jotform SMB Lead Qualification

Ingest Jotform submissions, classify lead quality with Gemini, and auto-route hot leads to sales. Confident leads hit Slack immediately; ambiguous ones get a clarifying email — no manual sorting needed.

Read the tutorial → · Download the code (zip)

Built with @reaatech/confidence-router, @reaatech/confidence-router-classifiers, @reaatech/structured-repair-core, @reaatech/agent-handoff-routing, @reaatech/llm-cost-telemetry, @reaatech/session-continuity on Google Gemini · 87 tests / 100% coverage.

Anthropic Document Pipeline for NetSuite SMB Invoice Processing

Extract invoice fields from PDFs and emails with Claude, then push vendor bills to NetSuite automatically. Semantic caching for recurring vendors, per-invoice cost caps, and automatic fallback to cheaper models keep spend predictable.

Read the tutorial → · Download the code (zip)

Built with @reaatech/llm-cache, @reaatech/llm-cache-adapters-redis, @reaatech/structured-repair-core, @reaatech/llm-cost-telemetry, @reaatech/agent-budget-engine, @reaatech/llm-router-core on Anthropic · 84 tests / 97.95% coverage.

AWS Bedrock Reliability Suite for Auth0 SMB Auth Operations

Wrap every Auth0 management API call with circuit breakers and idempotent retries. Secrets rotate automatically, and when authentication failures spike, Bedrock triggers a pre‑defined runbook and alerts Slack.

Read the tutorial → · Download the code (zip)

Built with @reaatech/circuit-breaker-core, @reaatech/idempotency-middleware, @reaatech/secret-rotation-core, @reaatech/secret-rotation-provider-aws, @reaatech/agent-runbook-agent, @reaatech/agent-runbook-alerts on AWS Bedrock · 94 tests / 95.09% coverage.

Mistral AI Voice Agent for After-Hours Customer Support

Deploy a voice receptionist that answers after-hours calls, understands natural speech, and books appointments — all with Mistral’s LLM, Deepgram STT, and ElevenLabs TTS. Session continuity keeps context across calls.

Read the tutorial → · Download the code (zip)

Built with @reaatech/voice-agent-core, @reaatech/session-continuity, @reaatech/voice-agent-telephony, @reaatech/voice-agent-stt, @reaatech/voice-agent-tts, @reaatech/agent-memory on Mistral · 110 tests / 95.76% coverage.

Mistral AI Document Pipeline for Shopify Tax Document Extraction

Pull tax IDs, totals, and line items from Shopify PDF invoices and receipts with Mistral’s OCR‑capable models. Structured repair keeps the output clean, and cost telemetry tracks per‑document spend.

Read the tutorial → · Download the code (zip)

Built with @reaatech/media-pipeline-mcp-doc-extraction, @reaatech/structured-repair-core, @reaatech/llm-cost-telemetry, @reaatech/confidence-router on Mistral · 79 tests / 98.78% coverage.

Anthropic Code Sandbox for SMB Data Cleansing Pipelines

Execute LLM‑generated data‑cleaning code inside an isolated sandbox. It automatically repairs malformed outputs, tracks costs per job, and supports idempotent retries — no risk of corrupting live data.

Read the tutorial → · Download the code (zip)

Built with @reaatech/structured-repair-core, @reaatech/confidence-router, @reaatech/llm-cost-telemetry, @reaatech/idempotency-middleware on Anthropic · 89 tests / 100% coverage.

Google Gemini Security Guardrails for SMB Healthcare PII Redaction

Detect and redact PHI from patient messages before they reach Gemini, helping small clinics stay HIPAA‑compliant. A guardrail chain sequences PII detection, redaction, and re‑identification across multi‑turn conversations.

Read the tutorial → · Download the code (zip)

Built with @reaatech/guardrail-chain, @reaatech/guardrail-chain-guardrails, @reaatech/llm-cost-telemetry, @reaatech/session-continuity, @reaatech/tool-use-firewall-core on Google Gemini · 72 tests / 99.60% coverage.

Instant lead response agent for fitness studios

Auto‑reply to inbound lead forms in under two seconds, matching chain studio speed. The agent classifies intent, routes through a mesh, and stores the interaction for follow‑up — no more lost prospects.

Read the tutorial → · Download the code (zip)

Built with @reaatech/agent-mesh, @reaatech/agent-mesh-router, @reaatech/agent-mesh-classifier, @reaatech/agent-memory, @reaatech/agent-memory-storage · 70 tests / 99.17% coverage.

Automated no-show and late-cancel enforcement for studios

Chase no‑shows and late cancellations automatically. It reads attendance records, applies policy, and sends LLM‑powered SMS via Twilio — reducing revenue leakage without manual follow‑up.

Read the tutorial → · Download the code (zip)

Built with @reaatech/agent-runbook, @reaatech/agent-runbook-agent, @reaatech/agent-runbook-runbook, @reaatech/agent-runbook-alerts, @reaatech/agent-runbook-health-checks · 100 tests / 99.47% coverage.

OpenAI Knowledge Agent for Confluence SMB Internal Wiki

A natural‑language Q&A bot that indexes Confluence spaces and delivers instant answers. It caches semantically similar questions, routes with confidence scores, and keeps session context for follow‑ups.

Read the tutorial → · Download the code (zip)

Built with @reaatech/confidence-router, @reaatech/session-continuity, @reaatech/llm-cache, @reaatech/agent-memory-core, @reaatech/agent-handoff, @reaatech/agents-markdown on OpenAI · 121 tests / 99.65% coverage.

Anthropic Multi-Agent Handoff for Freshdesk SMB Support Triage

Classify and route Freshdesk tickets to specialist AI agents with confidence‑based human fallback. A mesh gateway orchestrates the pool, and session continuity prevents lost context during handoffs.

Read the tutorial → · Download the code (zip)

Built with @reaatech/agent-handoff, @reaatech/agent-mesh-gateway, @reaatech/session-continuity, @reaatech/confidence-router, @reaatech/agent-handoff-compression, @reaatech/agent-handoff-routing on Anthropic · 109 tests / 99.58% coverage.

Azure AI Lead Intake for Calendly SMB Lead Qualification

An embedded chatbot that qualifies website visitors through natural conversation and books a Calendly meeting when they’re sales‑ready. Confidence scoring, session management, and webhook‑relay keep the pipeline smooth.

Read the tutorial → · Download the code (zip)

Built with @reaatech/confidence-router, @reaatech/session-continuity, @reaatech/agent-handoff, @reaatech/structured-repair-core, @reaatech/webhook-relay-core on Azure AI · 161 tests / 98.71% coverage.

AI Interview Coordinator for Boutique Recruiting Firms

Schedule interviews across candidates and panelists without the email back‑and‑forth. A voice‑first agent checks availability, creates Google Calendar events, and routes scheduling requests through an agent mesh.

Read the tutorial → · Download the code (zip)

Built with @reaatech/agent-mesh, @reaatech/agent-mesh-router, @reaatech/agent-handoff, @reaatech/agent-handoff-routing, @reaatech/agent-memory, @reaatech/agent-memory-retrieval · 117 tests / 97.32% coverage.

Automated Resume Rubric Scorer for Boutique Recruiters

Score 200 resumes per role against custom rubrics in minutes. An evaluation harness with golden trajectories, cost tracking, and a Redis‑backed cache turns a 10‑hour manual chore into a single POST request.

Read the tutorial → · Download the code (zip)

Built with @reaatech/agent-eval-harness-golden, @reaatech/agent-eval-harness-tool-use, @reaatech/agent-eval-harness-cost, @reaatech/agent-eval-harness-observability, @reaatech/llm-cache, @reaatech/llm-cache-adapters-redis · 121 tests / 98.81% coverage.

Azure AI MCP Server for SMB Business Intelligence

Deploy a multi‑tenant MCP server for AI agents to query BI dashboards. Per‑client rate limits, isolated tools, and OpenTelemetry tracing come built in — no bespoke multi‑tenancy code required.

Read the tutorial → · Download the code (zip)

Built with @reaatech/mcp-server-core, @reaatech/mcp-server-engine, @reaatech/multi-tenant-mcp-middleware, @reaatech/mcp-server-observability on Azure AI · 77 tests / 99.37% coverage.

Cohere Code Sandbox for Mixpanel Small Business Analytics

Ask natural‑language questions about your Mixpanel data and get budget‑constrained Python code executed in a sandbox. A budget engine, cost telemetry, and an LLM cache keep spending predictable.

Read the tutorial → · Download the code (zip)

Built with @reaatech/agent-budget-engine, @reaatech/llm-cost-telemetry, @reaatech/structured-repair-core, @reaatech/llm-cache on Cohere · 83 tests / 97.36% coverage.

vLLM Reliability Suite for NetSuite SMB Financial Operations

Enrich NetSuite records with a self‑hosted vLLM endpoint, guarded by circuit breakers and idempotency. The pipeline tolerates network failures, prevents duplicates, and traces every step to Langfuse.

Read the tutorial → · Download the code (zip)

Built with @reaatech/circuit-breaker-core, @reaatech/session-continuity, @reaatech/idempotency-middleware, @reaatech/agent-runbook-agent on vLLM · 123 tests / 98.15% coverage.

Perplexity Security Guardrails for Okta SMB Identity Protection

Intercept every AI‑driven Okta operation with a guardrail chain that detects PII, prompt injection, and policy violations. No raw identity data reaches the LLM — the chain stops it first.

Read the tutorial → · Download the code (zip)

Built with @reaatech/guardrail-chain, @reaatech/llm-cost-telemetry, @reaatech/structured-repair-core on Perplexity · 72 tests / 96.07% coverage.

OpenRouter Code Sandbox for SMB Financial Modeling

Let non‑technical analysts run Python financial models through a secure, budget‑aware sandbox. A firewall blocks dangerous calls, a router downgrades expensive models, and repair keeps outputs valid.

Read the tutorial → · Download the code (zip)

Built with @reaatech/confidence-router, @reaatech/tool-use-firewall-core, @reaatech/agent-budget-engine, @reaatech/structured-repair-core, @reaatech/llm-cost-telemetry, @reaatech/session-continuity on OpenRouter · 84 tests / 99.51% coverage.

OpenAI Agent Eval Harness for SMB Customer Support Quality

Automatically score support conversations on helpfulness, safety, and cost with an LLM‑as‑judge harness. Quality gates block deployments that regress below thresholds.

Read the tutorial → · Download the code (zip)

Built with @reaatech/agent-eval-harness-suite, @reaatech/agent-eval-harness-cli, @reaatech/agent-eval-harness-gate, @reaatech/agent-eval-harness-judge, @reaatech/llm-cost-telemetry on OpenAI · 92 tests / 96.11% coverage.

xAI Grok Secure Code Sandbox for SMB Data Pipelines

Execute AI‑generated code on your data without fear. A tool‑use firewall inspects every call, budgets cap spend per user, and structured repair fixes the LLM’s output before execution.

Read the tutorial → · Download the code (zip)

Built with @reaatech/agent-budget-engine, @reaatech/tool-use-firewall-core, @reaatech/tool-use-firewall-config, @reaatech/agent-budget-spend-tracker, @reaatech/structured-repair-core, @reaatech/agent-budget-pricing on xAI Grok · 117 tests / 95.32% coverage.

Databricks LLM Observability Suite for SMB AI Operations

Get end‑to‑end visibility into every LLM call on Databricks. OpenTelemetry GenAI semantics track token usage and cost, with ready‑made dashboards and alerts to prevent bill shock.

Read the tutorial → · Download the code (zip)

Built with @reaatech/otel-genai-semconv-core, @reaatech/otel-genai-semconv-openai, @reaatech/otel-genai-semconv-anthropic, @reaatech/otel-cost-exporter, @reaatech/llm-cost-telemetry on Databricks · 127 tests / 100% coverage.

Review Response Automator for Independent Restaurants

Generate on‑brand, empathetic replies to every Yelp, Google, or TripAdvisor review without staff effort. Sentiment analysis, guardrails, and agent memory keep the tone consistent and the response human‑like.

Read the tutorial → · Download the code (zip)

Built with @reaatech/a2a-reference-core, @reaatech/a2a-reference-client, @reaatech/a2a-reference-server, @reaatech/agent-handoff, @reaatech/agent-memory, @reaatech/guardrail-chain · 93 tests / 94.70% coverage.

Supplier Order Optimizer for Small Restaurant Groups

Auto‑calculate order quantities from current pars and rolling usage history. The agent reads spreadsheets, caches LLM results, and remembers order history so you don’t have to.

Read the tutorial → · Download the code (zip)

Built with @reaatech/a2a-reference-persistence, @reaatech/agent-memory-core, @reaatech/agent-memory-retrieval, @reaatech/agent-memory-storage, @reaatech/llm-cache, @reaatech/llm-cache-adapters-redis · 70 tests / 91.66% coverage.

Anthropic RAG Pipeline for Google Workspace SMB Email Knowledge Search

Ask plain‑English questions and get cited answers from your Gmail and Drive. Nightly ingestion builds embeddings, and quality gates catch retrieval regressions before users notice.

Read the tutorial → · Download the code (zip)

Built with @reaatech/rag-eval-core, @reaatech/rag-eval-gate, @reaatech/rag-eval-cost, @reaatech/rag-eval-dataset, @reaatech/llm-cache, @reaatech/session-continuity, @reaatech/llm-cost-telemetry on Anthropic · 127 tests / 97.01% coverage.

Vertex AI Lead Intake for Amazon SP‑API SMB Seller Acquisition

Replace static forms with a conversational agent that qualifies sellers, collects documents, and registers them via Amazon’s SP‑API. Budget‑aware orchestration prevents cost overruns.

Read the tutorial → · Download the code (zip)

Built with @reaatech/agent-mesh, @reaatech/agent-handoff, @reaatech/llm-cost-telemetry, @reaatech/agent-budget-engine, @reaatech/structured-repair-core on Vertex AI · 141 tests / 99.05% coverage.

Vercel AI Gateway Observability for SMB AI Agent Operations

Auto‑instrument every LLM call routed through Vercel AI Gateway with OpenTelemetry GenAI semantics. Cost aggregation, budget alerts, and a dashboard keep visibility without per‑provider code.

Read the tutorial → · Download the code (zip)

Built with @reaatech/otel-genai-semconv-core, @reaatech/llm-cost-telemetry, @reaatech/otel-genai-semconv-exporters, @reaatech/llm-cost-telemetry-aggregation, @reaatech/llm-cost-telemetry-calculator, @reaatech/llm-cost-telemetry-observability on Vercel AI Gateway · 93 tests / 97.33% coverage.

Vercel AI Gateway Agent Eval Harness for SMB Support Bots

An automated regression testing pipeline that replays golden conversations, scores responses with an LLM judge, and blocks CI if quality drops. Works against support bots using Vercel AI Gateway.

Read the tutorial → · Download the code (zip)

Built with @reaatech/agent-eval-harness-cli, @reaatech/agent-eval-harness-suite, @reaatech/agent-eval-harness-gate, @reaatech/llm-cost-telemetry, @reaatech/llm-cache, @reaatech/structured-repair-core on Vercel AI Gateway · 87 tests / 98.43% coverage.

Databricks RAG Pipeline for Insurance Policy Analysis

Query lengthy policy PDFs with natural language, backed by Databricks LLMs and pgvector. Semantic caching and context‑window planning keep answers accurate even with large documents.

Read the tutorial → · Download the code (zip)

Built with @reaatech/agent-memory-retrieval, @reaatech/agent-memory-storage, @reaatech/llm-cache, @reaatech/context-window-planner, @reaatech/structured-repair-core, @reaatech/llm-cost-telemetry on Databricks · 111 tests / 96.55% coverage.

vLLM Agent Mesh for E‑commerce Order Management

A multi‑agent system for handling order inquiries, shipment tracking, and returns. Intent classification, session continuity, and structured repair keep every interaction on track.

Read the tutorial → · Download the code (zip)

Built with @reaatech/agent-mesh, @reaatech/agent-mesh-registry, @reaatech/agent-mesh-router, @reaatech/agent-handoff, @reaatech/session-continuity, @reaatech/llm-cost-telemetry, @reaatech/structured-repair-core on vLLM · 165 tests / 99.42% coverage.

Auto‑Takeoff Agent for Small GC Bid Prep

Convert plan sets to bills of materials and subcontractor RFPs in minutes. OCR, LLM extraction, and agent meshing replace days of manual measurement and typing.

Read the tutorial → · Download the code (zip)

Built with @reaatech/media-pipeline-mcp-doc-extraction, @reaatech/agent-mesh, @reaatech/llm-cache, @reaatech/agent-budget-engine, @reaatech/a2a-reference-persistence, @reaatech/agents-markdown · 127 tests / 96.48% coverage.

Field Punch‑List Agent for GC Superintendents

Capture photos and voice memos on the jobsite, extract actionable items, and sync them directly to PM tools. Guardrails and agent memory keep the list accurate and traceable.

Read the tutorial → · Download the code (zip)

Built with @reaatech/voice-agent-core, @reaatech/media-pipeline-mcp-deepgram, @reaatech/agent-memory, @reaatech/webhook-relay-webhooks, @reaatech/guardrail-chain, @reaatech/a2a-reference-auth · 135 tests / 95.53% coverage.

Anthropic Lead Intake for Clio Small Law Firm Client Qualification

Qualify potential clients, capture case details, and create Clio matters through conversational intake. Confidence routing, session continuity, and cost tracking keep the pipeline lean.

Read the tutorial → · Download the code (zip)

Built with @reaatech/confidence-router, @reaatech/confidence-router-core, @reaatech/agent-handoff, @reaatech/agent-handoff-routing, @reaatech/structured-repair-core, @reaatech/session-continuity, @reaatech/session-continuity-storage-dynamodb, @reaatech/llm-cost-telemetry on Anthropic · 88 tests / 97.46% coverage.

AWS Bedrock Knowledge Agent for Buildium Tenant Self‑Service

Answer tenant questions about leases, payments, and maintenance by pulling live Buildium data. Semantic cache and structured repair keep responses fast and accurate.

Read the tutorial → · Download the code (zip)

Built with @reaatech/agent-memory, @reaatech/agent-memory-embedding, @reaatech/agent-memory-storage, @reaatech/session-continuity, @reaatech/session-continuity-storage-dynamodb, @reaatech/llm-cost-telemetry, @reaatech/llm-cache, @reaatech/llm-cache-adapters-qdrant, @reaatech/structured-repair-core on AWS Bedrock · 131 tests / 98.94% coverage.

Azure AI Agent Eval Harness for SMB Support QA

Automate quality gates for Azure AI‑powered support agents. The harness scores accuracy, tool correctness, and cost with an LLM judge and blocks regressions in CI/CD.

Read the tutorial → · Download the code (zip)

Built with @reaatech/agent-eval-harness-suite, @reaatech/agent-eval-harness-cost, @reaatech/agent-eval-harness-gate, @reaatech/agent-eval-harness-judge on Azure AI · 42 tests / 100% coverage.

Anthropic AI Runbook Automation for SMB Field Service Dispatching

Detect dispatch agent failures and execute pre‑defined runbooks automatically. Health checks, failure modes, and Slack alerts trigger only when automatic recovery fails.

Read the tutorial → · Download the code (zip)

Built with @reaatech/agent-runbook-agent, @reaatech/agent-runbook-runbook, @reaatech/agent-runbook-failure-modes, @reaatech/agent-runbook-health-checks, @reaatech/agent-runbook-alerts, @reaatech/agent-runbook-observability on Anthropic · 56 tests / 100% coverage.

LangChain Security Guardrails for SMB E‑commerce Support Bots

Add prompt injection defense, PII redaction, and content safety to existing LangChain agents without changing agent definitions. A configurable guardrail chain wraps every call.

Read the tutorial → · Download the code (zip)

Built with @reaatech/guardrail-chain, @reaatech/guardrail-chain-config, @reaatech/guardrail-chain-guardrails, @reaatech/guardrail-chain-observability on LangChain · 31 tests / 98.64% coverage.

OpenRouter Cost Control for SMB API Spend Management

Enforce daily budgets and automatically downgrade to cheaper fallback models when caps are at risk. Cost telemetry, budget engine, and router fallback work as a proxy in front of OpenRouter.

Read the tutorial → · Download the code (zip)

Built with @reaatech/llm-cost-telemetry, @reaatech/llm-cost-telemetry-calculator, @reaatech/agent-budget-engine, @reaatech/llm-cost-telemetry-aggregation, @reaatech/llm-cost-telemetry-exporters, @reaatech/llm-router-fallback on OpenRouter · 58 tests / 98.56% coverage.

Automated family inquiry to instructor match agent for tutoring centers

Instantly match incoming family inquiries to the best instructor. An LLM‑powered classifier scores every instructor against the inquiry, chooses the best fit, and dispatches via MCP.

Read the tutorial → · Download the code (zip)

Built with @reaatech/agent-mesh, @reaatech/agent-mesh-classifier, @reaatech/agent-mesh-router, @reaatech/agent-mesh-session, @reaatech/agent-mesh-observability, @reaatech/agent-mesh-mcp-server · 74 tests / 100% coverage.

Automated parent progress report generator for tutoring businesses

Generate consistent, data‑driven progress reports without waiting for instructor write‑ups. A document pipeline validates, analyzes, and formats reports from session data.

Read the tutorial → · Download the code (zip)

Built with @reaatech/agent-runbook, @reaatech/agent-runbook-agent, @reaatech/agent-runbook-runbook, @reaatech/agent-runbook-observability, @reaatech/agents-markdown, @reaatech/agents-markdown-reporter · 60 tests / 98.90% coverage.

Anthropic RAG Product Desc Generator for BigCommerce SMB Sellers

Generate SEO‑optimized product descriptions at scale using your BigCommerce catalog. A RAG pipeline with context‑window planning and structured repair works inside Claude’s token limits.

Read the tutorial → · Download the code (zip)

Built with @reaatech/session-continuity, @reaatech/context-window-planner, @reaatech/structured-repair-core, @reaatech/llm-cache on Anthropic · 157 tests / 98.11% coverage.

Azure AI Security Guardrails for Microsoft Entra ID SMB Access

Automate access‑request evaluation with PII detection, prompt injection blocking, and policy‑based approval. A guardrail chain that runs on Azure AI enforces compliance before provisioning.

Read the tutorial → · Download the code (zip)

Built with @reaatech/guardrail-chain, @reaatech/guardrail-chain-config, @reaatech/guardrail-chain-guardrails, @reaatech/guardrail-chain-observability, @reaatech/confidence-router-core on Azure AI · 93 tests / 97.38% coverage.

Google Gemini Knowledge Agent for eBay SMB Seller Support

Answer buyer questions by pulling live eBay data, with session continuity for multi‑turn conversations. Semantic search over listings and an LLM cache keep answers fast and accurate.

Read the tutorial → · Download the code (zip)

Built with @reaatech/session-continuity, @reaatech/context-window-planner, @reaatech/agent-memory, @reaatech/llm-cache, @reaatech/structured-repair-core on Google Gemini · 139 tests / 98.67% coverage.

vLLM Agent Quality Gate for On‑Prem SMB Support Bots

Add CI/CD quality gates to self‑hosted vLLM agents. Replay golden trajectories, score responses, and block deployment if quality drops below thresholds.

Read the tutorial → · Download the code (zip)

Built with @reaatech/agent-eval-harness-cli, @reaatech/agent-eval-harness-gate, @reaatech/agent-eval-harness-trajectory, @reaatech/agent-eval-harness-golden on vLLM · 64 tests / 100% coverage.

Mistral AI Spend Governance for Stripe SMB Subscription Billing

Enforce per‑customer LLM spend caps for Stripe billing workflows. When a tenant nears the soft cap, the router downgrades to cheaper Mistral models; at the hard cap, calls are blocked.

Read the tutorial → · Download the code (zip)

Built with @reaatech/agent-budget-engine, @reaatech/llm-router-engine, @reaatech/llm-cost-telemetry, @reaatech/agent-budget-spend-tracker, @reaatech/agent-budget-pricing on Mistral · 98 tests / 98.85% coverage.

Perplexity PII Shield for SMB E‑commerce Support Chat

Strip PII from support chat transcripts before sending to Perplexity for AI drafting. A guardrail chain redacts credit card numbers, emails, and phone numbers and logs every redaction for audit.

Read the tutorial → · Download the code (zip)

Built with @reaatech/guardrail-chain, @reaatech/guardrail-chain-guardrails, @reaatech/guardrail-chain-config, @reaatech/pi-bench-core, @reaatech/guardrail-chain-observability on Perplexity · 61 tests / 98.61% coverage.

Voice quote assistant for auto‑repair shops

Turn inbound “how much to fix X” calls into structured estimates without tying up your service advisor. Twilio, Deepgram STT/TTS, and the OpenAI Responses API generate quotes on the fly.

Read the tutorial → · Download the code (zip)

Built with @reaatech/voice-agent-core, @reaatech/voice-agent-stt, @reaatech/voice-agent-tts, @reaatech/voice-agent-telephony, @reaatech/agent-memory, @reaatech/llm-cache on OpenAI · 114 tests / 99.18% coverage.

Automated review response agent for tire shops

Never leave a negative review unanswered. The agent ingests reviews, runs guardrails, routes to the best LLM model, and generates empathetic, context‑aware replies referencing work orders.

Read the tutorial → · Download the code (zip)

Built with @reaatech/agent-memory-core, @reaatech/agent-memory-retrieval, @reaatech/llm-router-core, @reaatech/llm-router-strategies, @reaatech/guardrail-chain, @reaatech/guardrail-chain-guardrails · 91 tests / 96.03% coverage.

Anthropic Lead Intake for Pipedrive SMB Lead Enrichment

New Pipedrive leads are automatically enriched with company and intent data, fit‑scored, and routed to the right rep. Budget caps and session continuity keep the process controlled.

Read the tutorial → · Download the code (zip)

Built with @reaatech/agent-mesh, @reaatech/webhook-relay-core, @reaatech/confidence-router, @reaatech/agent-budget-engine, @reaatech/llm-cost-telemetry, @reaatech/session-continuity on Anthropic · 99 tests / 100% coverage.

Mistral AI Lead Intake for Marketo Lead Qualification & Routing

Ingest Marketo lead‑creation webhooks, enrich with Mistral AI, and route high‑intent prospects instantly. Confidence routing and cost tracking keep the pipeline efficient.

Read the tutorial → · Download the code (zip)

Built with @reaatech/agent-mesh, @reaatech/webhook-relay-core, @reaatech/confidence-router, @reaatech/session-continuity, @reaatech/llm-cost-telemetry, @reaatech/agent-budget-engine, @reaatech/llm-cache on Mistral · 101 tests / 97.45% coverage.

Browse all 53 tutorials →

Building blocks shipped

Agent eval harness

Two patches landed for the agent evaluation toolchain. agent-eval-harness-gate (0.1.1) restores CI and adds observability metrics tests, and agent-eval-harness-observability (0.1.1) upgrades the full OpenTelemetry stack to version 2.x. Both are backwards‑compatible.

Browse the family →

Idempotency middleware

The DynamoDB adapter bumped its AWS SDK dependencies to the latest major lines (3.50.0 → 3.1064.0) to keep the package current with the AWS ecosystem.

Browse the family →

Browse the full catalog at reaatech.com/products.