Field journal — category

ai articles

Log entries filed under ai, pulled from the operator's journal.

Filed under — ai15 entries

LOG 36Field journal23 Aug 2026

Technical illustration of a personal desktop workstation module docking into a larger automation orchestration machine through a single controlled gate.

Desktop AI Agents in 2026: What to Run Where

Desktop AI agents can now act on real files. Learn what belongs on a laptop, what belongs in orchestrated automation, and how to govern both in 2026.

8 min read

Read post

LOG 35Field journal22 Aug 2026

Technical atelier illustration of a mechanical lead intake system routing incoming enquiries through a scoring gate into three separate priority channels.

AI Lead Qualification Agents: The 2026 Build Guide

How to build an AI lead qualification agent in 2026: the rubric, structured output, CRM scoring, human handoff and the metric that proves it works.

9 min read

Read post

LOG 34Field journal21 Aug 2026

Technical atelier illustration of a modular MCP server component being inspected under a magnifying lens at a gate before entering a sealed business system.

MCP Server Security: How to Vet a Server in 2026

MCP server security starts before you connect. A consultant's checklist to vet servers, scope credentials, and keep a poisoned tool away from your data.

8 min read

Read post

LOG 32Field journal19 Aug 2026

Technical illustration of a sealed mechanical agent chamber driven by a scheduling timer and bounded by a calibrated spend gauge, representing Claude Managed Agents running business tasks on managed infrastructure.

Claude Managed Agents in 2026: When to Use Them

Claude Managed Agents in 2026: what the beta gives you, how token and session-runtime billing works, and when a workflow tool is the better call.

10 min read

Read post

LOG 31Field journal17 Aug 2026

Technical atelier illustration of a mechanical drafting arm assembling a component that passes through an inspection gate onto a running conveyor, representing AI coding agent output handed to a production runtime.

AI Coding Agents in 2026: What Ops Teams Can Build

AI coding agents changed in 2026. What ops teams can safely build with them, the three-question test before production, and when to use n8n instead.

8 min read

Read post

LOG 30Field journal16 Aug 2026

Technical illustration of a mechanical document-sorting system narrowing many archived files through filtering gates down to one selected card in a reading frame.

RAG for Business Knowledge Bases: What Works in 2026

RAG for business knowledge bases fails on retrieval, not the model. A 2026 build guide to chunking, reranking, permissions and measuring what works.

9 min read

Read post

LOG 23Field journal10 Aug 2026

Technical illustration of a modular drawer cabinet feeding one open module into a central mechanical reading head, representing an AI agent loading a single skill on demand.

Agent Skills for Business Operations: A 2026 Playbook

Agent Skills for business operations turn SOPs into instructions AI agents load on demand. The format, the real limits, and how to roll them out safely.

9 min read

Read post

LOG 22Field journal7 Aug 2026

Technical atelier illustration of a website drawn as a mechanical intake gateway sorting AI search crawlers, agents and training bots into three separate gated channels.

How to Make Your Business Website AI Agent Ready in 2026

An AI agent ready website in 2026 is a bot policy plus readable content. What Google, OpenAI, Anthropic and Cloudflare actually require, and what to skip.

8 min read

Read post

LOG 19Field journal27 Jul 2026

Technical atelier illustration of a stateless request-routing manifold sending one request into any of several identical interchangeable server modules, with a disconnected session-key component set aside.

The MCP 2026-07-28 Spec: What Changes and What Breaks

The MCP 2026-07-28 spec makes the protocol stateless and removes the initialize handshake. Here is what breaks, what it fixes, and how to plan your migration.

8 min read

Read post

LOG 16Field journal21 Jul 2026

Technical atelier illustration of a central orchestration hub routing signals to modular business components, representing an AI operating system for a small business.

The AI Operating System for Founders: A 2026 Guide

An AI operating system replaces scattered SaaS with one orchestration layer. Here's how founders build one in 2026: the stack, the steps, and the trade-offs.

7 min read

Read post

LOG 15Field journal17 Jul 2026

Technical illustration of a rotary telephone handset connected to a branching network of mechanical signal routing channels, representing an AI voice agent directing business calls.

AI Voice Agents for Small Business in 2026: A Buying Framework

AI voice agents for small business reached real maturity in 2026 — here's a practical framework for evaluating cost, compliance, and platform fit.

7 min read

Read post

LOG 14Field journal15 Jul 2026

Schematic illustration of a central knowledge-hub console routing tasks to small agent figures, with one path connecting to a separate gear-driven automation engine, representing Notion AI agents handing off work to a dedicated automation platform.

Notion AI Agents in 2026: What They Can Replace (and Can't)

Notion shipped External Agents and Workers in 2026. Here's what Notion AI agents actually replace for founders, and what still needs real automation.

7 min read

Read post

LOG 13Field journal14 Jul 2026

Technical illustration of a mechanical conveyor channel with a hand-shaped gate lever controlling an automated routing arm, representing a human approval checkpoint in an AI agent workflow

Human-in-the-Loop AI Agents: The 2026 Approval Framework

Human-in-the-loop AI agents don't need approval on every action. Here's the four-tier risk framework that keeps AI agent oversight tight in 2026.

6 min read

Read post

LOG 10Field journal6 Jul 2026

Technical illustration of a mechanical funnel filtering many labeled tool components down through a locking gate into one verified channel, representing controlled AI agent tool access.

AI Agent Tool Sprawl: How Anthropic's 2026 Upgrades Fix It

AI agent tool sprawl quietly inflates automation costs. See how Anthropic's 2026 MCP upgrades cut context bloat and add real access control.

7 min read

Read post

LOG 04Field journal26 May 2026

Technical atelier illustration of a central dispatch mechanism with multiple timed output channels radiating outward, representing automated follow-up pathways running on schedule without manual intervention

How Startup Teams Can Use AI Agents to Reduce Manual Follow-Ups

Learn how startup teams can use AI agents to reduce missed follow-ups, keep CRM records updated, and improve lead response without losing human control.

6 min read

Read post

All journal entries