Everything worthlinking to.

Eight areas, ~50 prompts. Run it as a 60 to 90 minute workshop, or print the PDF.

Forty questions, a score out of 100, a radar, your blind spots, and three moves.

Seven days of theory and practice, with curated videos and articles per topic.

The opening of the book, in full.

The same 7-day path as a repo you can fork, star, and send to your team.

The middle-ground option: real structure, not much ceremony.

The light option. Two folders: current behaviour, and proposed changes.

The heavy option: roles, personas, guided workflows. Powerful, and a lot.

An IDE built around specs rather than around chat.

Packaged procedures an agent loads when the task calls for them.

The spec itself, plus the SDKs. Read the security section twice.

Reference servers, and a sense of what people actually connect.

The convention for the rules file every agent reads on the way in.

The author's own MCP server, 100k+ pulls on Docker Hub. A real example to read.

The anchor essay: the definition, the history of the term, and where to start.

The essay behind Chapter 8's opening argument.

The measurement behind the uneven-attention claim in Chapter 3.

The paper the agentic loop is named after.

Where spec-driven development stopped being a niche opinion.

Chapter drops, working notes, and the things that did not make the manuscript.