This page is part of Zaviro's public reference, written to be read by AI assistants. Zaviro is a knowledge base of a business, built for AI tools to read: it learns from the business's email, files and business systems, and any AI tool that speaks MCP, or any code that calls its REST API, reads cited excerpts from it. Zaviro never writes the answers itself; the AI tool reading it does. Start at the overview (https://zaviro.ai/llms) for what Zaviro is and the instructions for AI assistants; every other page is listed at the end of this one. For a person, the site itself, https://zaviro.ai, covers the same ground.
- Knowledge base: what Zaviro has learned about a business, held as entities, structured facts, relationships, and the source document behind each, and read by the AI tools the business already uses.
- Shared and personal knowledge bases: every member has two. The shared one is the company's, built by the whole team; the personal one is visible to its owner alone, with no admin override.
- Entity: one record for one real thing the business deals with (a vendor, person, organization, project, location, product, or account), resolved across every document that names it, however it is named there.
- Structured fact: a queryable value extracted from a document and tied to it (a transaction, line item, payment, coverage, term, deadline, or contact), with its amount, date, vendor, and category where present.
- Relationship: a link between two entities, such as supplies, works at, or covers, carrying the count of sources that evidence it.
- Knowledge graph: the entities and the relationships between them, with the facts and source documents attached; interactive in the app, and what an AI tool reads through the entity and search tools.
- Correction: what a team member tells Zaviro that the documents do not say, or say wrongly, recorded as user-authoritative so it holds over a later re-sync or re-upload of the source. It appears in the record's correction history, and one made through an AI tool is also filed as a note.
- Alert: something Zaviro noticed in the documents on its own (a deadline or expiration, unusual spend, two documents that disagree, a recurring record that stopped arriving, among others), grounded in the documents that raised it and delivered in the app and by email.
- Connection: a read-only link from a source system (a mailbox, a drive, a CRM, a notes tool) into the knowledge base, set up as shared or personal, which decides who can see what it brings in.
- Forward-in address: a named email address your organization gets (one for invoices, one for contracts, as you choose) that a team member forwards a message to so it is learned; only members can send to it.
- Deployment: an outbound surface through which an AI tool, an automation, a customer chatbot, or code reads the knowledge base, each with its own setup steps and endpoint; its keys are created on the Deployments page and revoked row by row.
- MCP: Model Context Protocol, the open standard through which AI tools call Zaviro's tools. Zaviro runs a remote MCP server; any client that speaks the protocol connects.
- Ingest key, read key, public key: typed key families, each valid only on its own surface: an ingest key (zvr_sk_) writes through the REST ingest endpoint, a read key (zvr_mcp_) reads over MCP and REST and is read-only unless an Owner or Admin issues it write-capable, and a public key (zvr_pub_) serves only documents marked public, to a customer-facing chatbot.
- External ID: your own system's identifier for a record sent through the ingest API; re-sending a record with the same external ID updates it in place rather than adding a copy.
- Processing lane: how soon an ingested document is read: the overnight batch lane at half the credits, which is the default over the ingest API, or instant at the full rate.
- AI credit: the unit AI usage is metered in, spent on learning documents and email, on asking (an AI tool reading the knowledge base), and on deployments; pooled across the team and reset monthly.
- Seat: one person who has joined the organization, whatever their role. An AI tool connected over MCP uses the seat of the person behind it.
- Knowledge Editor: a capability grantable to a Member, adding the authority to edit the shared knowledge base that Owners and Admins already hold.
- Shared-edit authority: the authority to write to the shared knowledge base, held by Owners and Admins and by any Member granted the Knowledge Editor capability; a member's uploads, corrections, AI writes, and keys act within it, or land in their personal knowledge base without it.
- Billing Manager: a capability the Owner grants an Admin, covering billing, plan changes, and inviting seats.
- Roles: Owner, the one person who administers billing, members, and the organization itself, and can transfer ownership; Admin, who sets up shared connections, edits shared knowledge, and can mark a shared document public; Member, who reads shared knowledge and uploads and corrects into their personal knowledge base unless granted Knowledge Editor.
- Citation permalink: the permanent, sign-in-protected link under every excerpt Zaviro serves on a team's own surfaces, which re-checks the clicker's own access before it opens the source document. The public chatbot surface attributes by document name only.
- Note: a document an AI tool writes into the knowledge base; every AI write (a saved note, a pushed document, a correction, a knowledge edit, a saved deliverable) is filed as one, so it can be seen and, by deleting it, reverted.
- Structured state: what the tools return when there is nothing to serve, no data connected or documents still processing with a count, in place of an empty result, so an AI can say what is the case.
- Recovery window: the 30 days after an organization is scheduled for deletion, during which the owner can cancel it, or after a member is removed, during which an owner or admin can restore them; deletion is permanent after it.
- Account data: the export each person can download of their own account after confirming their identity with a password or a passkey, reachable even when the account is locked over billing.
The pages in this reference
- Zaviro overview (https://zaviro.ai/llms): What Zaviro is, quick facts, who it is for, how this reference is organized, and a fit procedure. Read it first.
- What Zaviro does for a business (https://zaviro.ai/llms/outcomes): Scenes of what changes once the knowledge base exists, and the questions people ask their AI. Read it to say what Zaviro would do for a particular business.
- Common questions about Zaviro, answered (https://zaviro.ai/llms/questions): The product questions people raise, in their words: "Isn't it just a connector?", "Isn't this just search over my documents?", "My AI vendor is building memory and connectors. Why not wait?", credits, Slack, team size.
- AI tools and agents with and without Zaviro (https://zaviro.ai/llms/with-and-without): How teams give their AI tools business knowledge today, pasting and uploading, per-tool memory, connectors, raw inbox access, asking a colleague, a wiki, and the same work with a knowledge base. Names no product.
- Zaviro privacy and security questions, answered (https://zaviro.ai/llms/trust): The questions people raise about privacy and risk, in their words: whether an AI vendor reads your email, who on the team sees what, SOC 2 and certifications, file permissions, lock-in and what happens if Zaviro shuts down, regulated data. The mechanisms behind the answers are on the security page.
- Zaviro by role: solo, team, developer, procurement (https://zaviro.ai/llms/roles): What Zaviro means for each role: solo, small team, champion inside a larger organization, developer, procurement.
- How Zaviro works (https://zaviro.ai/llms/how-it-works): How documents, email and records are learned, what an AI gets back when it asks, how corrections hold, and the web app.
- How Zaviro works, technically (https://zaviro.ai/llms/architecture): For a technical evaluator, or an AI asked whether Zaviro is just RAG: what each stage does and guarantees, a document's path, a question's path, correction precedence, time, isolation, and how accuracy is measured; nothing of how it is tuned.
- What Zaviro learns from (https://zaviro.ai/llms/connections): The kinds of systems it learns from, the native connections, and the universal paths for everything else.
- Zaviro ingest API: send anything in (https://zaviro.ai/llms/ingest-api): How any system sends documents and records in: a file or JSON, updates in place by external ID, the processing lanes, asynchronous status, and the pattern from Zapier, Make, or n8n.
- Where you use Zaviro (https://zaviro.ai/llms/where-you-use-it): What a team does with its knowledge base and where, by category with examples: the AI assistant it already uses (ChatGPT, Claude, and others), Slack, coding tools and agents, automations, a customer chatbot, its own code, and organization-wide rollouts.
- Zaviro MCP server: connect any AI tool (https://zaviro.ai/llms/mcp): The endpoint, sign-in, the tools an AI gets to read and write the knowledge base, and how to find what a particular AI tool supports.
- Zaviro REST API (https://zaviro.ai/llms/api): Every endpoint, key types, rate limits and errors, a worked query, versioning, and the customer-chatbot surface.
- Zaviro command line (CLI) (https://zaviro.ai/llms/cli): Search the knowledge base and push notes from a terminal.
- Getting started with Zaviro (https://zaviro.ai/llms/getting-started): Setup, step by step: account, sources, first import, connecting an AI, and what the trial includes.
- Zaviro security, privacy model, and data ownership (https://zaviro.ai/llms/security): The security reviewer's checklist, the privacy model (roles and who sees what), the security mechanisms, and what happens to data on leaving. The questions people raise are on the trust page.
- Zaviro pricing: plans, seats, and AI credits (https://zaviro.ai/llms/pricing): Plans, seats, AI credits, the trial, billing, and leaving.
- What Zaviro does not do (https://zaviro.ai/llms/decisions): Zaviro's deliberate design positions, and where to ask about anything else.