All flashcards

SharePoint's new prompt-first site builder. Takes a description of the site you want and scaffolds the landing page, lists, columns, and suggested views.

The approval pauses Ralph hits for persistent or cross-site changes. "Autonomous" doesn't mean unsupervised — the agent still asks for sign-off where it matters.

The max-iterations argument you pass when starting Ralph. Hard upper bound that prevents runaway loops. Your safety leash on autonomous runs.

The sample-data prompt pattern: fix some fields ("assign all to me") and vary the rest ("randomize the choices") to get useful test rows quickly.

A follow-up "make this prettier" prompt that applies basic styling — levels, status icons, emojis. Not brand-aware unless a style skill is loaded.

A skill that traps SharePoint AI in a self-iterating loop. Given a goal and a max-iteration cap, Ralph plans, acts, grades itself, and retries until it passes.

The five steps Ralph does each iteration — the acronym behind the loop's name. Reason on goal, act on plan, look at result, probe for gaps, harden what works.

Asking the AI to make N realistic-looking rows with some fields pinned and others randomised. Lets you design views against representative data.

A single paragraph that names the domain, the artefacts (site, lists), and the column schema. The Agentic Builder produces the whole shell from it.

The plan Ralph writes on iteration 1 and grades against thereafter. Defines what "done" means — list, views, data, styling — and is the loop's stopping rubric.

Ralph and the style definition skill loaded together. Ralph drives the iteration; the style skill shapes each output — neither knows about the other.

Filter views the builder proposes based on columns you defined (e.g. status → Approved Sessions, Pending Review). Free starting point you can edit.

The video's structure — easy (Agentic Builder), moderate (manual prompting + voice), advanced (Ralph self-iterating loop). Same goal, three levels of automation.

A dedicated prompt that builds a named filtered view ("show me approved sessions"). Runs after the list and data exist — separate from styling.

Speaking long natural-language prompts to SharePoint AI. Fast once the structured schema work is done and you're into paragraph-shaped requests.