Agents & Automation tag | eYssen AI

16 Articles

Agents & Automation ×

Case Study: How We Built Our Own Agent

Krisztian Eyssen, Krisztian Eyssen

Case Study: How We Built Our Own Agent

Close to twenty AI tools running side by side, and whatever one of them learned, the next had never heard of. That is where EYAS came from. This post is not a feature list: it shows the decisions, the trade-offs we said out loud, and the one measurement that threw a finished-looking solution out of the window.

Advanced Business owner Case study EYAS Agents & Automation

Sep 13, 2026

Self-Hosted Personal Agents and Their Risks

Krisztian Eyssen, Krisztian Eyssen

Self-Hosted Personal Agents and Their Risks

The security policy of the most widely used open personal agent lists what does **not** count as a vulnerability. This is on that list: **"prompt injection, if it does not bypass policy, authentication, approval, the sandbox or a tool boundary."** And this: **"anyone who can operate an agent can make it do anything that agent is able to do."** This is not a flaw in the system — this is the system. This article is about what you need to know about that before you start it up.

Intermediate Everyone Explainer Agents & Automation

Sep 13, 2026

Testing and Monitoring Your Agents

Krisztian Eyssen, Krisztian Eyssen

Testing and Monitoring Your Agents

In a survey of 1,300 practitioners, **39.8%** of respondents ran offline evaluation and **32.5%** ran it online — while everyone named **quality** as the main obstacle to getting into production, more than twice as often as cost or security. Everybody logs. Measuring whether the agent is doing a good job is something most people cannot do. This article is about how you can.

Advanced Developer How-To Agents & Automation

Sep 13, 2026

Multi-Agent Systems: When More Agents Help

Krisztian Eyssen, Krisztian Eyssen

Multi-Agent Systems: When More Agents Help

Five real multi-agent systems were measured in the same study. The best one scored **84.8%**, the worst **13.3%** — and in the middle of the field sit two systems that got **three quarters of the tasks wrong**. The difference did not come from the model. It came from the pattern. This article is about **which four patterns pay back the cost of coordination, and which two do not.**

Advanced Developer Explainer Agents & Automation

Sep 13, 2026

Scheduled and Background Agents

Krisztian Eyssen, Krisztian Eyssen

Scheduled and Background Agents

One vendor's documentation describes its own interface like this: **"a green status means the session started and exited without an infrastructure error. It does not mean the task succeeded."** That one sentence sums up the entire problem with background agents. And according to one measurement, **more than a third** of failures are exactly this: the agent **does not know when to stop**.

Advanced Business owner How-To Agents & Automation

Sep 13, 2026

Browser and Computer-Use Agents: Real Uses

Krisztian Eyssen, Krisztian Eyssen

Browser and Computer-Use Agents: Real Uses

In April 2024 the best model solved **12.24%** of computer-use tasks; a human solved **72.36%**. In August 2026 the best number is **85.96%**. So the capability is there — the question is the cost. And it does not grow linearly, it grows **with the square of the step count**: 6.7 times more steps means a **34 times bigger bill**. The biggest turnaround, though, is somewhere else entirely: the agent that **writes code instead of clicking** is cheaper and more accurate at the same time.

Intermediate Business owner Tool review Agents & Automation

Sep 13, 2026

Agent Memory: What to Store and What to Forget

Krisztian Eyssen, Krisztian Eyssen

Agent Memory: What to Store and What to Forget

Memory is the weakest agent capability there is. On the best independent measurement, **the average of 109 models fails to reach 30% on any memory strategy** — the very same models score 86–91% on simple tool calls. That measurement also shows that **there is no universally good memory architecture**. And there is a second reason for caution: memory is an **attack surface** — in one study an attacker planted a persistent false memory with a **98.2% success rate**, using nothing but questions.

Advanced Developer Explainer Agents & Automation

Sep 13, 2026

Subagents and Parallel Work

Krisztian Eyssen, Krisztian Eyssen

Subagents and Parallel Work

The most quoted number says a multi-agent setup is **90.2% better** than a single-agent one. The same source also writes that it uses **fifteen times as many tokens**, and that **80% of the performance difference is explained by token consumption alone**. And a 2026 paper showed it formally: **at an equal token budget, the single-agent system wins.** The value of splitting a task is not parallelism — it is **clean context**.

Advanced Developer How-To Agents & Automation

Sep 13, 2026

Permissions: The Least-Privilege Agent

Krisztian Eyssen, Krisztian Eyssen

Permissions: The Least-Privilege Agent

Three things are needed for a data leak: **private data**, **untrusted content** and **an outbound channel**. Any two of them on their own are safe — all three together are not. And this is not theory: according to a study of 7,973 live servers, **40.55% of them expose their tools with no authentication**.

Advanced Business owner How-To Agents & Automation

Sep 13, 2026

Human in the Loop: Where to Put the Approval

Krisztian Eyssen, Krisztian Eyssen

Human in the Loop: Where to Put the Approval

The human who approves is not a neutral safety layer. In a 2026 randomised clinical trial, **44 doctors, after twenty hours of AI literacy training**, still performed **14 percentage points worse** whenever the AI handed them a wrong recommendation. Training does not protect against this. What does: **few approval points, but real ones.**

Intermediate Business owner How-To Agents & Automation

Sep 13, 2026

Agent Reliability Math: Why Twenty Steps Fail

Krisztian Eyssen, Krisztian Eyssen

Agent Reliability Math: Why Twenty Steps Fail

0.98 to the twentieth power = **66.8%**. That one multiplication explains the feeling that an agent is "almost right, but it always breaks something somewhere". And the reverse is harsher still: **at twenty steps, a 90% end-to-end success rate demands 99.475% per-step accuracy.** Not 90%. This whole topic lives in the gap between those two numbers.

Intermediate Business owner Explainer Agents & Automation

Sep 13, 2026

No-Code Automation with AI: Which Platform?

Krisztian Eyssen, Krisztian Eyssen

No-Code Automation with AI: Which Platform?

Price is not what decides, **the billing unit is**. Five large platforms, five different units — and the same daily automation with twenty steps consumes **900** units on one and **18,000** on another. That is a twentyfold difference for exactly the same work. Plus one correction that many comparisons get wrong: **n8n is not open source.**

Beginner Business owner Comparison n8n Agents & Automation

Sep 13, 2026

1
2

Follow Us

Archives

Tags

Audience

Everyone Office worker Marketer Creative Business owner Developer

Category

AI Basics Safety & Risk Regulation & Policy Infrastructure & Chips Business & Deals Prompting Models & Market Writing & Content Images & Design Video & Audio Marketing & SEO Coding & Development Work & Productivity Agents & Automation Data, RAG & Self-Hosted AI Business, Law & Security

Format

How-To Explainer Comparison Tool review Recipe Template Troubleshooting Case study Glossary News

Level

Beginner Intermediate Advanced

Tool

OpenRouter Perplexity HeyGen ElevenLabs Suno Obsidian MCP n8n Ollama EYAS

Others

Odoo