Skip to content

Data Readiness, 11 Building Blocks, and a Redesigned Build Phase

The Business-First AI Framework gets its biggest skill update since the Organizational Lens — a new deep-dive dimension, expanded building block vocabulary, a portable Design Prompt, and a restructured Build phase.

Data Readiness Probing (6-Question Framework)

Section titled “Data Readiness Probing (6-Question Framework)”

The Deconstruct step now uses a 6-question framework, adding Data Readiness as the sixth dimension alongside discrete steps, decision points, data flows, context needs, and failure modes. For every step’s inputs, the skill now probes access (can AI reach the data?), interpretability (is the format AI-friendly?), and persistence (does this knowledge exist as a durable artifact, or is it stuck in someone’s head?). The Context Shopping List output now includes AI Accessible? and Readiness Notes columns, and the Design skill extracts these flags to inform step classification and produces a Data Readiness Summary in the Building Block Spec.

The framework vocabulary expands from 7 to 11 building blocks, organized by the three-layer model: Intelligence (Model, Context, Memory, Project), Orchestration (Prompt, Skill, Agent), and Integration (MCP, API, SDK, CLI). The design skill now maps every step against the full set, ensuring workflows that need API integrations, SDK code, CLI automation, or persistent memory get those blocks identified during Design rather than discovered mid-build.

The monolithic building-workflows skill has been decomposed into three focused skills — designing-workflows (architecture decisions, autonomy assessment, step classification), constructing-workflows (artifact generation from the approved spec), and running-workflows (run guide creation). Each can be invoked independently, and the Business-First AI plugin is now at v6.0.2.

The Design step is now available as a standalone skill. Install it on any platform to run the full Design phase — assess autonomy, choose orchestration, classify steps, and produce a Building Block Spec.

The AI Workflow Examples plugin (v1.1.0) adds two demo skills for the Agentic AI course: extracting-article-insights and drafting-linkedin-posts — a two-step content repurposing pipeline that extracts key points from a source article and drafts a LinkedIn post from the results.

The GitHub setup guide now includes a Create a Repository step, walking students through creating their first repo from the command line.