Weekly recap, May 25, 2026 – May 31, 2026

26 step-by-step AI tutorials landed this week, covering everything from voice-powered field service dispatch to expense report extraction with xAI Grok. Each one includes a downloadable zip and a walkthrough you can follow in an afternoon. Here’s the full list.

New tutorials

Databricks Security Guardrails for SMB Data Pipelines

Add PII redaction, prompt injection defense, and content policy enforcement to your Databricks model-serving pipelines without retraining any models. This pluggable security layer sits between your users and Databricks endpoints, scanning both incoming prompts and model outputs, and logging all violations to Langfuse.

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 Databricks · 111 tests / 100% coverage.

Mistral AI Lead Intake for Clio Legal Client Onboarding

Automatically capture new client leads via web chat and document uploads, detect duplicates, and create contacts and matters in Clio’s REST API. Mistral powers a conversational qualifier that asks intake questions, while hybrid RAG indexes past leads to flag duplicates and ingested documents are processed for key details.

Read the tutorial → · Download the code (zip)

Built with @reaatech/hybrid-rag, @reaatech/hybrid-rag-retrieval, @reaatech/hybrid-rag-embedding, @reaatech/hybrid-rag-ingestion, @reaatech/agents-markdown, on Mistral · 115 tests / 100% coverage.

Anthropic Lead Intake for SMB Home Services Bookings

Capture and qualify leads from web forms for plumbers, electricians, and HVAC businesses using Anthropic’s Claude. Attachments like photos or PDFs are processed, structured service details extracted, and high-confidence leads automatically routed to a booking webhook while ambiguous ones are flagged for review.

Read the tutorial → · Download the code (zip)

Built with @reaatech/confidence-router, @reaatech/structured-repair-core, @reaatech/agent-handoff, @reaatech/llm-cost-telemetry, @reaatech/session-continuity, on Anthropic · 105 tests / 99.27% coverage.

Perplexity Agent Mesh for SMB Competitive Intelligence

Deploy a mesh of AI agents that continuously monitor competitors and market shifts, using Perplexity for web-grounded research. Specialist agents track pricing, news, and features; a classifier categorizes incoming intel; and the mesh hands off tasks between agents with session continuity.

Read the tutorial → · Download the code (zip)

Built with @reaatech/agent-mesh, @reaatech/agent-mesh-classifier, @reaatech/agent-mesh-session, @reaatech/agent-mesh-router, @reaatech/agent-handoff, @reaatech/llm-cost-telemetry, @reaatech/confidence-router, on Perplexity · 85 tests / 100% coverage.

xAI Grok Reliability Suite for SMB AI Operations

Proactively monitor, diagnose, and self-heal your AI agent operations with an automated CLI toolchain powered by xAI Grok. Generate operator runbooks with health checks, trip circuit breakers on anomalous error rates, and repair malformed structured outputs before they break your pipelines.

Read the tutorial → · Download the code (zip)

Built with @reaatech/agent-runbook, @reaatech/circuit-breaker-core, @reaatech/agent-runbook-alerts, @reaatech/agent-runbook-health-checks, @reaatech/circuit-breaker-agents, @reaatech/structured-repair-core, on xAI Grok · 101 tests / 94.49% coverage.

Mistral AI Lead Intake for Housecall Pro Field Service Bookings

Convert inbound phone calls into structured Housecall Pro job bookings with an AI voice agent. Twilio handles telephony, Deepgram transcribes, ElevenLabs synthesizes speech, and Mistral AI extracts booking details. A budget engine caps per-call spend.

Read the tutorial → · Download the code (zip)

Built with @reaatech/voice-agent-core, @reaatech/create-voice-agent, @reaatech/voice-agent-mcp-client, @reaatech/voice-agent-telephony, @reaatech/agent-budget-engine, @reaatech/structured-repair-core, @reaatech/session-continuity, on Mistral · 105 tests / 99.54% coverage.

Anthropic Document Pipeline for QuickBooks Online Invoice Automation

Extract invoice details from PDFs, emails, and scans, then sync them to QuickBooks Online to eliminate manual data entry. Claude does the structured extraction; hybrid RAG indexes past invoices to improve accuracy; and a budget engine controls costs.

Read the tutorial → · Download the code (zip)

Built with @reaatech/hybrid-rag, @reaatech/hybrid-rag-ingestion, @reaatech/hybrid-rag-evaluation, @reaatech/hybrid-rag-embedding, @reaatech/agent-budget-engine, on Anthropic · 51 tests / 95.86% coverage.

Vertex AI Multi-Agent Handoff for ServiceTitan Dispatch Automation

Triage field service requests via web, SMS, or voice transcript, then schedule appointments and assign technicians in ServiceTiton automatically. A multi-agent mesh uses Vertex AI for intent classification, confidence gating, and handoffs to specialist dispatch agents.

Read the tutorial → · Download the code (zip)

Built with @reaatech/agent-mesh, @reaatech/agent-mesh-classifier, @reaatech/agent-mesh-confidence, @reaatech/agent-handoff, @reaatech/agent-handoff-routing, on Vertex AI · 79 tests / 97.44% coverage.

xAI Grok Knowledge Agent for Buildium Property Management Queries

Answer tenant and property manager questions about leases, payments, and maintenance schedules using an AI agent connected to Buildium data. Hybrid RAG indexes lease agreements and work orders in a pgvector database, Grok generates responses, and session continuity preserves conversation context.

Read the tutorial → · Download the code (zip)

Built with @reaatech/hybrid-rag, @reaatech/agent-mesh, @reaatech/llm-router-core, @reaatech/session-continuity, @reaatech/llm-cost-telemetry, @reaatech/hybrid-rag-ingestion, on xAI Grok · 116 tests / 99.28% coverage.

xAI Grok Cost Control for SMB Customer Support Agents

Prevent runaway AI spending by enforcing per-tenant daily budgets and fallback routing for Grok-powered support agents. When budgets tighten, requests automatically downgrade to a cheaper OpenAI model. Real-time cost telemetry and gateway-level rate limiting keep spending predictable.

Read the tutorial → · Download the code (zip)

Built with @reaatech/agent-budget-engine, @reaatech/llm-router-core, @reaatech/llm-router-fallback, @reaatech/llm-cost-telemetry, on xAI Grok · 100 tests / 99.55% coverage.

Perplexity Agent Eval Harness for SMB AI Quality Assurance

Run continuous, automated evaluations of your customer-facing AI agents using Perplexity as a neutral LLM judge. The CLI pipeline pulls golden test cases from your repo, scores responses, gates prompt promotions based on quality thresholds, and streams results to Langfuse.

Read the tutorial → · Download the code (zip)

Built with @reaatech/agent-eval-harness-suite, @reaatech/agent-eval-harness-judge, @reaatech/agent-eval-harness-golden, @reaatech/prompt-version-control, @reaatech/classifier-evals, @reaatech/agents-markdown-linter, on Perplexity · 122 tests / 99.68% coverage.

Ollama Agent Mesh for Small Business Workflow Automation

Orchestrate a fleet of local LLM specialists that handle email triage, CRM updates, and report generation without sending data to the cloud. Everything runs on your own hardware through Ollama; confidence routing decides whether to act, clarify, or escalate; and session continuity maintains context across multi-turn tasks.

Read the tutorial → · Download the code (zip)

Built with @reaatech/agent-mesh, @reaatech/agent-mesh-router, @reaatech/confidence-router, @reaatech/session-continuity, @reaatech/agent-handoff, @reaatech/llm-router-core, on Ollama · 139 tests / 97.19% coverage.

AWS Bedrock Document Pipeline for AppFolio Lease Data Extraction

Extract lease data from PDFs and scanned images, then push structured records to AppFolio. Unpdf and Tesseract OCR handle text extraction, Bedrock (Claude) does structured extraction, and structured repair fixes malformed JSON before writing to AppFolio. A tool-use firewall requires human approval for destructive actions.

Read the tutorial → · Download the code (zip)

Built with @reaatech/structured-repair-core, @reaatech/llm-cost-telemetry, @reaatech/agent-budget-engine, @reaatech/tool-use-firewall-core, on AWS Bedrock · 97 tests / 100% coverage.

OpenAI Voice Agent for OpenTable Reservation Management

Let diners manage OpenTable reservations by phone using natural conversation. A voice agent answers calls via Twilio, transcribes with Deepgram, classifies intent with OpenAI, and synthesizes speech with Cartesia. The agent calls OpenTable’s REST API to create, modify, or cancel reservations.

Read the tutorial → · Download the code (zip)

Built with @reaatech/voice-agent-core, @reaatech/voice-agent-telephony, @reaatech/voice-agent-stt, @reaatech/voice-agent-tts, @reaatech/voice-agent-mcp-client, @reaatech/session-continuity, @reaatech/structured-repair-core, on OpenAI · 127 tests / 98.96% coverage.

Vertex AI Code Sandbox for QuickBooks Financial Modeling

Ask conversational questions about your QuickBooks data and get safe, sandboxed code execution. Vertex AI generates Python, an e2b sandbox runs it securely, and a tool-use firewall with human approvals ensures code doesn’t go rogue. Spend controls and cost telemetry keep you within budget.

Read the tutorial → · Download the code (zip)

Built with @reaatech/session-continuity, @reaatech/tool-use-firewall-core, @reaatech/tool-use-firewall-approvals, @reaatech/llm-cost-telemetry, @reaatech/structured-repair-core, on Vertex AI · 159 tests / 98.91% coverage.

Mistral AI Security Guardrails for Square Payment Dispute Resolution

An AI agent drafts responses to Square payment disputes while guardrails prevent unauthorized actions and redact PII. Mistral analyzes evidence, drafts responses, and a chain of guardrails filters PII, toxicity, and prompt injection before requiring human approval to submit. A budget engine caps monthly spend.

Read the tutorial → · Download the code (zip)

Built with @reaatech/guardrail-chain, @reaatech/guardrail-chain-guardrails, @reaatech/tool-use-firewall-approvals, @reaatech/llm-cost-telemetry, @reaatech/agent-budget-engine, on Mistral · 90 tests / 100% coverage.

Ollama Reliability Suite for On-Prem SMB Agent Operations

Keep local Ollama agents running 24/7 with automatic circuit breaking, runbook generation, session durability, and repair of broken structured outputs. The circuit breaker prevents hammering a downed model, runbooks document incident playbooks, and session continuity preserves conversation state across restarts.

Read the tutorial → · Download the code (zip)

Built with @reaatech/circuit-breaker-core, @reaatech/agent-runbook-agent, @reaatech/session-continuity, @reaatech/structured-repair-core, on Ollama · 198 tests / 100% coverage.

xAI Grok Voice Lead Intake for Small Business

A phone-based lead capture agent that qualifies callers using xAI Grok and syncs to your CRM. It answers calls, transcribes with Deepgram, uses Grok for understanding, classifies intent with a confidence router, and maintains session state with Redis for multi-turn conversations.

Read the tutorial → · Download the code (zip)

Built with @reaatech/voice-agent-core, @reaatech/mcp-server-core, @reaatech/session-continuity, @reaatech/voice-agent-telephony, @reaatech/voice-agent-mcp-client, @reaatech/session-continuity-storage-redis, @reaatech/confidence-router, on xAI Grok · 107 tests / 99.13% coverage.

Perplexity Security Guardrails for SMB Financial Queries

Safely use Perplexity’s search-augmented AI for financial Q&A with real-time PII redaction, harmful-content blocking, and audit logging. A guardrail chain routes every query through financial-specific rules: PII detection, topic enforcement, and toxicity filtering, with all violations logged to Langfuse.

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/llm-cost-telemetry, on Perplexity · 72 tests / 98.62% coverage.

xAI Grok Expense Report Extraction for SMB Finance

Automatically extract line items, totals, and merchant names from scanned receipts and invoices using xAI Grok’s vision model. Ingest PDFs and images, OCR them, pass text to Grok for structured extraction, repair malformed JSON, and export results to CSV files stored in S3.

Read the tutorial → · Download the code (zip)

Built with @reaatech/media-pipeline-mcp-doc-extraction, @reaatech/structured-repair-core, @reaatech/classifier-evals, @reaatech/llm-cost-telemetry, on xAI Grok · 99 tests / 98.99% coverage.

OpenAI Voice Agent for ServiceTitan Field Service Dispatch

Handle phone dispatch for home service businesses, converting voice instructions into ServiceTitan work orders automatically. The agent answers calls, transcribes speech, classifies intent, and creates work orders via ServiceTitan’s REST API with session continuity across callbacks.

Read the tutorial → · Download the code (zip)

Built with @reaatech/voice-agent-core, @reaatech/create-voice-agent, @reaatech/voice-agent-mcp-client, @reaatech/voice-agent-simulator, @reaatech/voice-agent-telephony, @reaatech/confidence-router, @reaatech/session-continuity, on OpenAI · 111 tests / 99.36% coverage.

Vertex AI Invoice Extraction for SMB Accounting

Turn stacks of invoices and receipts into clean QuickBooks transactions with Vertex AI document parsing and structured repair. Extract structured data from PDFs, repair malformed LLM output, route high-confidence fields to QuickBooks, and flag low-confidence items for human review.

Read the tutorial → · Download the code (zip)

Built with @reaatech/structured-repair-core, @reaatech/confidence-router, @reaatech/confidence-router-classifiers, @reaatech/llm-cost-telemetry, @reaatech/llm-cost-telemetry-calculator, on Vertex AI · 89 tests / 96.62% coverage.

Cohere AI Spend Control for Budget-Conscious SMBs

Enforce daily AI spend limits, automatically downgrade to cheaper Cohere models when budgets get tight, and get real-time cost dashboards. Telemetry wraps every LLM call, the router dynamically selects models, and the budget engine enforces hard caps per tenant.

Read the tutorial → · Download the code (zip)

Built with @reaatech/llm-cost-telemetry, @reaatech/llm-router-core, @reaatech/llm-router-engine, @reaatech/llm-cost-telemetry-aggregation, @reaatech/llm-cost-telemetry-calculator, @reaatech/agent-budget-engine, on Cohere · 130 tests / 97.09% coverage.

Google Gemini Lead Intake for Salesforce SMB Sales

Automatically extract and qualify leads from email attachments and web forms, pushing structured records to Salesforce. Gemini parses text from PDFs and images, a confidence router classifies leads, and hybrid RAG deduplicates against existing records. PII is scrubbed, and costs are tracked.

Read the tutorial → · Download the code (zip)

Built with @reaatech/hybrid-rag, @reaatech/confidence-router, @reaatech/agent-budget-engine, @reaatech/llm-cache, @reaatech/classifier-evals, on Google Gemini · 103 tests / 100% coverage.

Meeting notes to action items with Slack followup

Turn Zoom transcripts into assigned tasks and automatically follow up via Slack for 7 days. The pipeline extracts action items, assigns owners, and sends reminders — ensuring nothing falls through the cracks after client calls.

Read the tutorial → · Download the code (zip)

Built with @reaatech/agent-memory, @reaatech/agent-memory-extraction, @reaatech/agent-memory-retrieval, @reaatech/agent-handoff, @reaatech/agent-handoff-routing, @reaatech/agents-markdown, on agnostic · 66 tests / 94.42% coverage.

Automated triage of customer feedback into product roadmap items

Ingest support tickets, NPS comments, and sales call feedback, then cluster and prioritize them into a weekly product roadmap report. The evaluation harness scores and gates the output, ensuring only high-impact suggestions reach the product team.

Read the tutorial → · Download the code (zip)

Built with @reaatech/agent-eval-harness-suite, @reaatech/agent-eval-harness-judge, @reaatech/agent-eval-harness-types, @reaatech/agent-eval-harness-observability, @reaatech/agent-eval-harness-cost, @reaatech/agent-eval-harness-gate, on agnostic · 93 tests / 96.25% coverage.

Browse all solutions →