CORAS

Getting Data into CORAS.ai

Three ways to bring data into CORAS.ai — and when to use each one.

Before Gary can analyze your data, surface insights, or run scenarios, that data needs to live somewhere in CORAS.ai. There are three ways to get it there. Each one serves a different purpose and produces a different result. Understanding which method fits your situation saves time and produces better outputs from Gary.

Why This Matters

  • Gary's outputs are only as good as the data he's working from. Knowing how your data got into CORAS.ai — and which method is right for your use case — is the foundation of effective Gary conversations.

  • The three methods aren't mutually exclusive. Most organizations use all three. Structured program data comes in through ingestion. Reference documents go into the Library. One-off files get uploaded directly in Gary.

  • The method you choose determines who can see the data, how long it persists, and whether Gary can work from it across all conversations or just one.


The Three Methods


1. Central Ingestion

For structured data that needs to live in your SmartLists permanently.

Central Ingestion is how you bring spreadsheets, CSVs, and other tabular data into CORAS at scale. You upload a file, map its columns to CORAS fields, and import the records into a node and list. Once ingested, that data is available to Gary, SmartBoards, SmartLanes, and every other CORAS feature that reads from list data.

Use this when:

  • You have structured data in a spreadsheet or CSV that belongs in a SmartList

  • You're onboarding a new program, portfolio, or dataset into CORAS

  • You need data to persist across sessions and be available to your whole team

  • You want Gary to analyze that data as part of his standard list context

What to know:

  • Ingestion is an admin-level action

  • Records ingested here are visible to anyone with access to that node

  • Data stays in CORAS permanently until removed by an admin

  • Gary can analyze ingested data any time you point him at the right list

See Central Ingestion and Multi-Node Ingest for step-by-step instructions.


2. The Library

For reference documents and files your whole team needs access to in Gary.

The Library is your organization's document store in CORAS. Files uploaded to the Library become System Documents — automatically available in Gary's data context panel for any user with access to that library. Manuals, policy documents, reference guides, reports — anything your team needs to bring into a Gary conversation without attaching it to a specific list item.

Use this when:

  • You have reference documents that multiple users need to pull into Gary conversations

  • You want files available without requiring users to upload them individually each session

  • You're managing documents that don't belong to a specific list item but are relevant across workflows

What to know:

  • Library files are visible to everyone in the tenant — don't upload sensitive or restricted files here

  • Files need to be analyzed before Gary can work from them — look for the Analyze button next to the file name in the Sources panel

  • Once analyzed, Library files surface automatically in Gary's context panel under System Documents

  • Users select which Library files to include in a given conversation — they aren't attached by default

See Data Sources and Context for how Library files appear in Gary conversations.


3. In the Gary Window

For one-off files you need Gary to work from in a single conversation.

When you need Gary to analyze a specific document for the conversation you're in right now — a draft report, a data export, a policy memo — you upload it directly in the Gary chat window. The file is available for that session only and is visible only to you.

Use this when:

  • You have a document relevant to one specific conversation

  • The file doesn't need to be shared with your team or persist across sessions

  • You want Gary to extract insights, summarize, or analyze a document alongside your list data

What to know:

  • Supported file types: PDF, Word, Excel, .txt, .md, JSON

  • Files uploaded this way are Conversation Documents — they disappear when the conversation ends

  • Only you can see files uploaded directly in Gary

  • You can drag and drop files directly into the text box or use the + icon at the bottom left of the Gary window

See Gary AI — Getting Started for the full upload workflow.


Which Method Should I Use?

  • Structured data that belongs in a SmartList = Central Ingestion

  • Reference documents the whole team needs in Gary = Library

  • A one-off file for this conversation only = Gary Window

  • Data that needs to persist and be analyzed by Gary over time = Central Ingestion

  • A document attached to a specific list item = Attach to the item - it surfaces as a System Document automatically


Get Help

Contact your Customer Success Manager or visit the Customer Resource Center for additional support with data ingestion and Gary configuration.