Daily recap for May 23, 2026

Six new tutorials landed today, covering everything from self-hosted financial analysis to voice agents for order checking. Each includes a step-by-step guide and a zip of the working code.

New tutorials

Ollama Code Sandbox for SMB Financial Analysis

Run Python and SQL queries for financial modelling entirely on-premises with this self-hosted sandbox. Built on Ollama and Daytona, it keeps sensitive data in-house while @reaatech packages enforce spend limits, recover from sandbox failures, and repair malformed generated code before execution.

Read the tutorial → | Download the code (zip)

Built with @reaatech/agent-budget-engine, @reaatech/circuit-breaker-agents, @reaatech/llm-cache, on Ollama · 56 tests / 95.27%.

OpenAI Invoice Extraction for Xero SMB Accounting

Pull invoice data from PDFs and images directly into Xero using OpenAI’s vision models. High-confidence extractions post automatically; the rest get flagged for review. Hybrid RAG retrieves similar past invoices to improve accuracy, and a budget engine keeps LLM spend in check.

Read the tutorial → | Download the code (zip)

Built with @reaatech/hybrid-rag, @reaatech/hybrid-rag-embedding, @reaatech/agent-budget-engine, @reaatech/confidence-router, @reaatech/agent-eval-harness-golden, on OpenAI · 123 tests / 100.00%.

Cohere Voice Agent for Shopify Order Status Checks

Let customers call to ask “Where’s my order?” and get a spoken reply from Shopify, with Cohere handling the conversation. Circuit breakers gracefully handle Shopify API glitches, and an agent handoff passes complex cases to Zendesk. A per-call budget prevents cost overruns.

Read the tutorial → | Download the code (zip)

Built with @reaatech/agent-mesh, @reaatech/agent-handoff, @reaatech/circuit-breaker-agents, @reaatech/agent-budget-engine, on Cohere · 111 tests / 100.00%.

OpenAI Multi-Agent Handoff for E-commerce Order Handling

Three specialist OpenAI agents — for status, modifications, and cancellations — work together under Mastra orchestration to manage customer requests. A response cache speeds up common queries, an idempotency layer prevents duplicate refunds, and a per-conversation budget engine stops runaway costs.

Read the tutorial → | Download the code (zip)

Built with @reaatech/agent-handoff, @reaatech/agent-mesh, @reaatech/llm-cache, @reaatech/agent-budget-engine, @reaatech/idempotency-middleware, on OpenAI · 109 tests / 97.26%.

Anthropic Document Pipeline for SMB Lease Abstraction

Extract critical dates, rent amounts, and clauses from lease PDFs and DOCX files using Claude, with a retrieval-augmented clause library to boost accuracy. Confidence scoring routes low-certainty results to human review; high-certainty ones auto-populate an abstract. Budget limits and circuit breakers keep the pipeline predictable.

Read the tutorial → | Download the code (zip)

Built with @reaatech/hybrid-rag, @reaatech/confidence-router, @reaatech/agent-budget-engine, @reaatech/circuit-breaker-agents, on Anthropic · 146 tests / 99.70%.

Mistral AI Agent Reliability Suite for SMB Customer Support

Wrap your Mistral-powered support agents in per-tool circuit breakers, retry middleware, and automated incident playbooks. When a downstream API like your CRM or return portal slows down, the suite isolates the failure so the agent keeps working. Health checks generate Kubernetes probes automatically.

Read the tutorial → | Download the code (zip)

Built with @reaatech/circuit-breaker-agents, @reaatech/agent-runbook-health-checks, @reaatech/agent-runbook-incident, @reaatech/idempotency-middleware, @reaatech/agent-runbook-runbook, on Mistral · 63 tests / 98.51%.

Browse all solutions →

Browse the full catalog at reaatech.com/products.