Skip to content
LE LAB X

Public reading guide

Scope your first agent

Separate context, execution and verification before delegating.

For turning an automation idea into a bounded task.

Before starting: Be able to describe the task, its inputs and the expected result.

An ordered selection, not the complete catalogue. Articles are dated; full text remains at the source.

  1. 1 / 3 ·

    Everything you wanted to know about Hermes Agent

    Hermes Agent is an open-source AI agent that orchestrates tools, the web and files while retaining what it learns across sessions. More than a chatbot, it is a complete framework with persistent memory, reusable skills and an integrated research pipeline.

    Look for: Which agent capabilities fit your need?

  2. 2 / 3 ·

    Planka + Hermes Kanban: turning a board into a shared workspace for humans and agents

    Planka + Hermes Kanban turns a board into a shared workspace for humans and AI agents. Cards become work units that anyone can read, edit and execute, with better coordination, durable context and clear handoffs.

    Look for: How can you divide the work and track its result?

  3. 3 / 3 ·

    You don't need $30,000 in Claude Code credits: take GPT Pro beyond the browser

    Codex prepares the context, Oracle sends a controlled bundle to GPT‑5.6 Sol in ChatGPT Pro, and Codex reviews it before coding and testing. Roles stay separate: Pro reasoning on one side, Codex execution on the other—without the OpenAI API or sending the repository blind.

    Look for: Which boundaries should remain between review and execution?

Your next step

Choose a reversible task, limit its access and write down how you will verify the result.

All articles on this topic

The members-only wiki remains separate from these public readings.