liveBook · Manning

Markdown

As explained in Vibe Engineering

Markdown is a lightweight markup language for formatting plain-text documents.

Markdown is a lightweight markup language for formatting plain-text documents. In agent workflows, it carries externalized state such as specifications, plans, conventions, and project knowledge across context windows. section 3.5.3

Markdown also creates an integration boundary: generated content may be readable to people but unusable to software until its formatting is removed. SQL generation, for example, requires stripping fenced Markdown before execution. section 6.2

Use Markdown deliberately rather than treating it as accumulated project memory:

  • Externalize architecture, schemas, conventions, organizational policies, historical decisions, and operational knowledge in structured files. section 3.5.3
  • Expect orchestration overhead and additional failure modes when Markdown files become part of multi-agent workflows. section 3.5.3
  • Clean generated Markdown fences when downstream systems require executable SQL rather than formatted prose. section 6.2

Externalized project state

A project can preserve accumulated intelligence in a .planning/ directory, skill files, CLAUDE.md, and AGENTS.md, giving fresh agent sessions immediate access to shared context. section 3.5.3

Figure 3.11 The context pyramid: five interdependent layers of the Context Fabric. Each layer rests on the ones below it - a gap low in the pyramid undermines everything above, no matter how complete the upper layers are.

The context pyramid shows how Markdown-held knowledge can span system, code, organizational, historical, and operational context, with gaps in lower layers weakening higher layers. section 3.5.3

How Vibe Engineering covers it

How deep it goes
Shows operational uses, file examples, context layers, orchestration costs, and SQL cleanup, but does not detail Markdown syntax.
Examples
Uses planning directories, agent instruction files, context-layer artifacts, and fenced SQL responses as concrete examples.
What you can do after
Apply Markdown to externalize agent context and clean generated SQL before passing it to executable systems.
How it connects
Connects Markdown with context externalization, AI agents, prompt construction, SQL generation, and the OpenAI API.

Also explained in

Compare how 2 books explain Markdown →

Related concepts

This explainer was generated by Manning's knowledge graph from the text of Vibe Engineering. The section links point to the parts of the book it draws on.