> ## Documentation Index
> Fetch the complete documentation index at: https://docs.experio.cloud/llms.txt
> Use this file to discover all available pages before exploring further.

# How Experio Works

> The components of the platform, what each one does, and how a question becomes a cited answer

## The short version

Experio connects to a client's document libraries and business systems and reads what is there. It turns that content into a **knowledge graph**: one connected record of the things the business cares about, such as clients, projects, people, skills, contracts and obligations, and how they relate.

Users ask questions in chat, or run agents that produce documents. Experio answers from the graph and cites the files and records behind each answer.

```mermaid theme={null}
flowchart LR
    subgraph Sources["Client sources"]
      DOC["Document libraries<br/>SharePoint, Box, Google Drive"]
      TAB["Business systems<br/>HR, CRM, finance exports and APIs"]
    end
    subgraph Build["Build the graph"]
      ING["Document pipeline<br/>parse, classify, extract, match"]
      STR["Structured pipeline<br/>data mappings"]
      ENR["Enrichment rules"]
    end
    G[("Knowledge graph<br/>entities, relationships,<br/>lineage")]
    subgraph Use["Use the graph"]
      CHAT["Chat assistants"]
      AF["Agent flows"]
      MCP["MCP / other tools"]
    end
    DOC --> ING --> G
    TAB --> STR --> G
    G --> ENR --> G
    G --> CHAT
    G --> AF
    G --> MCP
```

Everything in the middle column is driven by configuration that the implementation team agrees with the client: the [ontology](/implementation/key-concepts#ontology), [taxonomies](/implementation/key-concepts#taxonomy), [artifact types](/implementation/key-concepts#artifact-type), [data mappings](/implementation/key-concepts#data-mapping), [matching strategies](/implementation/key-concepts#matching-strategy) and [enrichment rules](/implementation/key-concepts#enrichment-rule). The software does the reading. The configuration tells it what to look for and how to connect it.

## The components

You don't need to operate these services, but it helps to know they exist when you read job status or talk to Experio support.

| Component | What it does | Where you see it |
| - | - | - |
| **Web application** | The chat interface for users and the admin panel for configuration. It stores all configuration and starts scans. | `/` and `/admin` |
| **Reader** | Scans the folders set on a [data source](/implementation/key-concepts#data-source) and records each file it finds. | **Process > Jobs** |
| **Coordinator** | Moves each file through the pipeline, one step at a time, and retries failures. | File status on **Process > Jobs** |
| **Downloader** | Fetches each file from the storage provider. | File status: download |
| **Parser** | Turns PDF, Word, PowerPoint, Excel and other files into text. It can use OCR for scanned documents when the data source enables it. | File status: parse |
| **Classifier** | Decides which artifact type a file is, with a confidence score. Low-confidence files go to human review. | File status: classify; **Process > Conflict Resolution** |
| **Ingestion** | Extracts entities and relationships using the ontology, matches them to existing records, and writes them to the graph. | File status: ingest |
| **Structured data** | Loads CSV, Excel and JSON files and API responses through a data mapping. | **Process > Jobs** |
| **Enrichment** | Runs enrichment rules over nodes already in the graph. | **Model & Define > Enrichment Rules** |
| **Graph evaluation** | An AI judge that audits a sample of ingested documents for extraction errors. | **Process > Graph Evaluation** |
| **Flow runner** | Runs [flows](/implementation/key-concepts#flow): scheduled chains of scan, ingestion and enrichment jobs. | **Process > Flows** |
| **Message queue** (RabbitMQ) | Hands work between the services. | Not visible to admins |
| **Application database** (PostgreSQL) | Holds configuration, job state, review queues and ontology revisions. | Not visible to admins |
| **Graph database** (Neo4j or FalkorDB) | Holds the knowledge graph itself, including embeddings used for similarity search. | **Administer > System Settings** (`GRAPH_PROVIDER`) |
| **AI models** | Large language models used for classification, extraction, matching decisions, enrichment and chat. Each job can use a different model. | **AI & Agents > Model Configurations** |

<Note>
  Each Experio deployment is dedicated to one client. The client's data, graph and configuration are not shared with other clients.
</Note>

## Building the graph

There are two ways in, and they work differently. [How Data Gets to the Graph](/implementation/data-to-graph) covers both in detail.

* **Documents** (proposals, contracts, resumes, reports): an AI model reads each file and extracts only what the artifact type asks for. It matches what it finds against existing records, using similarity, AI review and human review.
* **Tables** (exports from HR, CRM and finance systems, or API responses): a data mapping tells Experio exactly which column becomes which entity, attribute or relationship. Matching is by exact key.

Both land in the same graph, on the same entity types. The goal of an implementation is that "Lakeshore Health" from the CRM, from a folder name and from the text of a contract all end up as **one** Client node, with every source recorded in its [lineage](/implementation/key-concepts#lineage).

## Using the graph

### Chat assistants

When a user asks a question, the assistant:

1. **Routes** the request. It decides whether this is a question about internal data, a request that also needs external tools such as email or CRM, a follow-up to the previous answer, or a job for an agent flow.
2. **Resolves names** in the question to graph records. "Lakeshore" becomes the Client node *Lakeshore Health*. This uses exact matches and the similarity (vector) indexes set per entity type on the ontology's **Indexes** tab.
3. **Plans a query.** It works out which graph paths answer the question, writes a graph query from the ontology and taxonomies, and checks the query before running it.
4. **Gathers context** from the matched records and their neighbours, and optionally from document text.
5. **Writes the answer** with citations. Users can open each citation and trace its lineage back to the file or row it came from.

Citations respect [access control](/implementation/key-concepts#access-control). A user who can't see a contract won't see it cited.

### Agent flows

An [agent flow](/implementation/key-concepts#agent-flow) is a multi-step AI workflow built on a visual canvas. A typical flow collects inputs, pulls data from the graph, has a model rank or draft something, pauses for a human to approve, and produces a Word, PowerPoint or Excel file. Users launch flows from chat. The assistant can also start one when a request matches the flow's description.

### Other ways in

* **Document templates**: branded PowerPoint and Word output from chat, through `/generate`, the wand button, or by asking for a document type.
* **MCP server**: users can call Experio from Claude Desktop, Cursor and other MCP clients with a personal API key.
* **Integrations**: per-user Google Workspace and Slack connections, so an assistant can combine graph answers with the user's email, calendar or channels.

## Where the configuration lives

The admin panel (`/admin`) groups its pages by stage of the work. Most implementation effort is in the first three sections.

| Sidebar section | Pages you'll use in an implementation |
| - | - |
| **Model & Define** | Taxonomies, Ontology, Compatibility, Artifact Types, Data Mapping, Enrichment Rules, Template Categories, Document Templates |
| **Connect** | Connectors, Data Sources |
| **Process** | Flows, Flow Executions, Matching Strategies, Conflict Resolution, Jobs, Graph Evaluation |
| **AI & Agents** | Model Configurations, AI Instructions, Agent Configuration (assistants), Personas, Copilot, Agent Flows, Agent Inbox, MCP Servers |
| **Observe** | Startup Health, Logs, Scaling, Usage Statistics |
| **Administer** | Client Configuration, Access Control, SSO Configuration, User Profile Management, System Settings |

Every admin page has an **Admin Copilot** (<kbd>⌘</kbd>/<kbd>Ctrl</kbd> + <kbd>J</kbd>). It answers questions from this guide and the Admin Guide, and on some pages it can draft changes for you to review.

## Next

<CardGroup cols={2}>
  <Card title="How Data Gets to the Graph" icon="route" href="/implementation/data-to-graph">
    The document and structured pipelines step by step, and every setting that affects accuracy.
  </Card>

  <Card title="Key Concepts" icon="book" href="/implementation/key-concepts">
    Definitions of ontology, taxonomy, artifact type, data mapping and the rest.
  </Card>
</CardGroup>