Foundation
The four context sections every agent reads first: company, ecosystem, product and ideal customer.
App route: /brand_context/foundation
Foundation is the base layer of your context: what your company does, the market it sits in, what you sell, and who you sell it to. Agents read these four sections before they research a topic, score a page, or draft a sentence, which makes this the highest-leverage screen in Context. Everything downstream inherits whatever is written here, including the mistakes.
Autopilot drafts all four during setup from public research. Your job is not to write them from scratch: it's to correct the places where public research got you wrong, and to mark each section Ready when it reads true.
Doing this rather than looking it up? The Tutorial walks this screen step by step in 3.1 · Review your context.
Reference
The screen lists four sections, each with its description, who last edited it, when, and its status.
| Section | What it holds |
|---|---|
| Company Overview | What your company does, who it serves, and what makes it different |
| Ecosystem | The market, categories, and players your brand operates within |
| Product & Features | Your products, their core features, and the value each delivers |
| Ideal Customer | The companies and buyers you're the best fit for |
Last edited shows a person's name when a human has touched the section, and Agent when it is still exactly as autopilot drafted it. That column is the quickest read on the screen: anything still attributed to Agent has never been checked by anyone.
Status is Ready or Empty, and it isn't something you set: it's derived from whether the section's document has any content. There's no "reviewed" state and no partial state: a section autopilot drafted and nobody has read still shows Ready.
Ready is a hard prerequisite, not a progress bar
This is the most important thing on the screen, and it's easy to read the wrong way round. That same has-content condition is a prerequisite on roughly twenty agent workflows. When a required section is empty, those workflows refuse to start: they don't run with thinner context and produce a weaker result.
What's gated on an empty Company Overview or Product & Features includes persona generation, competitor discovery and briefing, taxonomy generation, information architecture, topic-cluster bootstrapping, topic discovery, writing-profile compilation, opportunity generation, content briefs, and content writing. Setup autopilot halts rather than continuing past the gap.
A smaller set of workflows do degrade gracefully: they include the section if it exists and carry on without it if not: so the behaviour isn't perfectly uniform. But the default is refusal.
Two consequences worth holding onto:
- Ready does not mean checked. It means non-empty. An autopilot draft nobody has read is Ready, and is already shaping every downstream agent.
- Empty is not neutral. It's the state that stops production. If work has mysteriously stalled: no opportunities appearing, briefs not generating: an empty foundation section is the first thing to check.
Personas, Competitors and Taxonomy have their own completion tests, which read Done rather than Ready. Personas and Taxonomy check that records exist; Competitors is stricter, wanting at least one competitor and no competitor left un-enriched.
Ideal Customer is the one people under-edit. It reads fine as generic B2B prose, and that generic version quietly steers page targeting for months. It's worth being specific enough that the description would be wrong for your closest competitor.
Why it works this way
Last updated at October 20, 2018