> ## Documentation Index > Fetch the complete documentation index at: https://cofounder.appeeky.com/llms.txt > Use this file to discover all available pages before exploring further. # Company Knowledge Base > What you tell the Co-founder about your business — profiles, business context, and the Library — and which parts reach every briefing. The Vault is what the agent **learned**. The knowledge base is what you **told it**. Keeping the two separate is the point: a belief the agent inferred from store data should never quietly outrank a fact you stated about your own company. Everything here is optional. An owner who fills in nothing gets exactly the briefing they got before this existed — the renderers return an empty string when there is no content, so nothing is added to the prompt. ## Four layers | Layer | What | Scope | Reaches the prompt | | ------------------------- | -------------------------------------------------------------------------------- | ------------------ | ------------------------------------------ | | **Company profile** | Structured fields — stage, ICP, positioning, brand voice, hard constraints | Your whole account | **Always**, verbatim | | **App profile** | Per-product fields — monetization, known weaknesses, roadmap focus, do-not-touch | One app | **Always** for that app's runs | | **`business_context.md`** | The prose version: what you're building, who it's for, what you already tried | Your whole account | **Always**, truncated at a heading if long | | **Library** | Everything else — departmental notes, research, playbooks | Account or one app | **Retrieved** when relevant | Below all of this sits the [Vault](./vault) — facts and episodes, scored and evicted. Untouched by this feature and still the learning loop. ### Precedence ``` Profiles > business_context.md > Library > Vault facts ``` When a fact contradicts a profile field, the profile wins and a `contradicts` edge is written in the memory graph. The reverse never happens on its own: the agent can only *propose* a change to your context, never make one. See [Knowledge Patches](../capabilities/knowledge-patches). ### Why company and app are separate A founder with five apps has one company and five products. Brand voice, business model, and company-wide constraints are shared; ICP, pricing, and *"do not touch the app name"* are per-product. Merging them would force you to either repeat yourself five times or write a profile too vague to change any recommendation. Runs are per-app, so each one is injected `company + this app`. ## What the fields are for Two fields are worth more than the rest combined: * **`constraints`** (company) — *"Never recommend removing the free tier."* Rendered as a blocking rule; the agent will not propose an action that breaks one. * **`doNotTouch`** (app) — *"app name", "the free tier", "MX pricing".* Turns *"please stop suggesting this"* from a weekly chat correction into structure. Both stay **switched off while the profile is an unreviewed draft**. A hallucinated constraint blocking real work is the worst failure this feature can have, so a rule only becomes binding once you confirm it. ## Token budget Always-on context has to stay bounded or it displaces the store data the briefing is actually about. | Block | Budget | On overflow | | ----------------------------------- | ----------------------------------------------- | --------------------------------------------------------------------------------------------------------- | | Company profile | \~350 tokens | Fields are short by construction | | App profile | \~250 tokens | Same | | `business_context.md` + pinned docs | \~1,200 tokens (4,800 chars, split across them) | Truncated at the last heading that fits; the agent is told which sections were dropped and can fetch them | | Retrieved Library docs | \~900 tokens, max 4 docs | Lowest-similarity docs dropped first | | **Total always-on** | **≤ 2,500 tokens** | | Truncation always happens at a **heading boundary**, never mid-sentence — a severed clause reads to the model as a complete thought, which is worse than a missing section. At most **3 documents** can be pinned always-on; a fourth would only shrink the other three. To give a single app its own always-on context, pin a document and scope it to that app: pinned documents are read on every briefing *for the apps that can see them*. `business_context.md` itself stays account-wide. Every run logs a `company_context_injected` event with the rendered size, so budget creep is visible before it degrades a briefing. ## You don't start from a blank form Roughly eight minutes after you connect an app, the Co-founder drafts your profiles and a first `business_context.md` from evidence it already has: | Source | Fills | | ------------------------------------------------------- | ----------------------------------------------------------------------- | | Your workspace listing (imported store metadata) | Product one-liner, value prop, differentiators, category, launch status | | Tracked competitors | Positioning set | | Existing Vault facts (`brand_voice`, `target_audience`) | Seed values, marked as prior | | Public reviews | Target user, core jobs, known weaknesses | | Website (seller URL) | Company name, mission, ICP copy | Reading your own workspace beats re-fetching the store: it is richer than a lookup response, works the same for Google Play, and can't be rate-limited. A live store lookup happens only when the workspace has no description. Three guards keep the draft honest: * **It never overwrites you.** Any profile you already confirmed is skipped, and `business_context.md` is left alone if it contains a single word of your own prose. * **Weak evidence writes nothing.** The model is told to return null rather than guess. Fewer than two populated fields → no draft, because a review banner over an empty form is worse than no banner. * **Vault-only evidence writes nothing.** Re-deriving a profile from facts the agent already injects just restates them in a second block. Until you confirm it, the draft **is still injected** — prefixed *"(unconfirmed draft — treat as low confidence)"*. A wrong-but-editable draft beats no context, and the marker stops the agent over-committing to it. ## Where to edit it | Surface | What | | ----------------------- | ------------------------------------------------------------------------------------------------------------------------------- | | **Settings → Business** | Profile fields, the draft review banner, confirm-all, pending patch queue, and a preview of the literal text injected into runs | | **Library** | `business_context.md` and every other document, in a markdown editor | | **Chat** | *"What do you know about us?"* renders the context block verbatim | | **MCP / CLI** | `workspace_search`, `workspace_list`, `workspace_read`, `workspace_write`, `workspace_delete` | | **API** | [Company](../api/company), [Documents](../api/documents), [Knowledge Patches](../api/knowledge-patches) | `business_context.md` is edited in the **Library only**, deliberately. Two editors over one document means two autosave loops racing on the same body, and the loser silently loses a paragraph. ## Related docs * [Library](../capabilities/library) — documents, scope, retrieval, and the agent's document tools * [Knowledge Patches](../capabilities/knowledge-patches) — how the agent proposes changes to any of this * [Memory](./memory) — the full memory model including the Vault * [Second Brain](../capabilities/second-brain) — ingesting outside content (articles, Notion, files) into the Vault