Agent entry

Low-noise paths into the garden.

These files are generated from the same article folders that power the human site. Agents should start with the compact index, then fetch article packets by slug.

Discovery feeds

The index files are intentionally compact. They expose article IDs, topics, maturity, claim counts, source counts, and packet locations without forcing an agent to parse prose.

agents/index.json agents/index.jsonl llms.txt graph/nodes.json graph/edges.json

Article packets

Each article has a human page, an agent Markdown brief, and a JSON packet with claims, sources, relationships, review state, and a content hash.

seed / published

Discovery feeds

Article packets

Tool Use: When the Model Calls Something Outside Itself

Retrieval-Augmented Generation: Looking Things Up Before Answering

Reasoning Models: Slower Thinking, Better Checks?

Prompt Engineering: Instruction Design, Not Magic Words

Prompt Caching: Reusing Stable Context

Planning and Reflection: How AI Breaks Down and Revises Work

Multi-Agent Systems: When More Than One AI Worker Is Involved

Memory vs Context: What Should Survive the Conversation?

Loops vs Goals: The Difference Between Repetition and Direction in AI Agents

Long-Running Sessions: Keeping AI Work Coherent Over Time

Fine-Tuning: Teaching a Model a Narrower Behavior

Evaluations: How We Know an AI Workflow Improved

Context Management: What the AI Sees Right Now

AI, De-Mystified: A Field Guide to Modern AI Terminology

Agents: Goal-Directed AI Systems That Use Tools

AI Delegation Orchestration: A Series on Durable Agent Work

Commitment Boundaries in High-Stakes Domains

Capability Contracts for Agent Networks

Long-Running Delegations: How Agents Can Work for Hours Without Losing the Plot

Control Loci, Not Human Managers: An Agent-Native Routing Model

The Operator Cockpit Problem: Why More Traces Are Not Enough

The Delegation Record: A Schema for Consequential AI Work

From Conversation to Delegation: Why AI Work Needs a Durable Record

Listening to the Firehose: Can Voice-First, Two-Way Audio Become a Legitimate Assistive Medium?

You Do Not Need to Learn AI First: A 5-Minute Conversation Recipe

Beyond the First Conversation: Advanced Questions for New AI Agent Users

From Agent Swarms to Agent Control Planes

The Long Human Road to AI: A Reader’s Guide to Season 1

Learning Machines: Statistics, Neural Networks, and the Data Turn

The Human Road Through AI: Labor, Institutions, Governance, and Meaning

Foundation Models and the Return of General-Purpose AI Systems

From Formal Logic to Computation: The Mathematical Road to AI

The Birth of AI: Dartmouth, Symbolic Systems, and Early Optimism

Before Machines: Calculation, Automata, and the Dream of Mechanical Reason

Winters, Expert Systems, and the Cost of Overpromising Intelligence

Agentic Commerce and the Product Truth Layer

The Future of Publishing Is Agent-Auditable Research

Roadmap packet

Aura Knowledge Roadmap