Knowledge Base

A living technical reference where material claims are connected to evidence. The feed answers "what changed" — the knowledge base answers "what is this technology, how does it work, and what should you consider before adopting it."

Pages

Subject hubs

Linked claims

Model Context Protocol

The open protocol for connecting AI models to tools, data sources, and external systems.

AI Agents

Autonomous and semi-autonomous AI systems that plan, use tools, and execute multi-step tasks.

Core concept What is an AI agent?

AI Security

Threat models, vulnerabilities, and defences for AI systems — prompt injection, tool poisoning, and more.

Risk & security Prompt injection, tool poisoning, and confused-deputy risks

Comparisons TRACE

Side-by-side evaluations of AI models, tools, and approaches with evidence-based recommendations.

Definitions TRACE

Clear explanations of AI concepts, protocols, and terminology.

How-To Guides TRACE

Practical guidance on securing, evaluating, and building with AI systems.

Security Advisories TRACE

Security risks, threat models, and defensive practices for AI systems.

security advisory How should an AI coding agent be secured? Review due

How the knowledge base works

Not a wiki. Edits flow through review before publication. No open public editing.

Evidence-first. Every material claim must have at least one evidence record. No unsourced assertions.

Versioned. Every change produces a version record with a change summary. No silent rewrites.

Connected to the feed. Intelligence events can propose updates to knowledge pages. Knowledge pages surface related events.

Inference-labelled. Editorial inferences are clearly distinguished from directly sourced claims.

Read the full methodology →