Skip to content

Integrations ​

Bring documents in from the tools you already use, give your chat assistant extra tools through MCP, and use your Scrapalot library from your own AI agent.

Plans

External connectors, MCP integrations and API keys are part of the Pro plan and above. See Pricing.

Connectors ​

Connectors import documents from an external source into a collection. Open Knowledge Stacks, select a collection, and go to the Connectors tab. Pick a connector, fill in its details (the form shows setup steps for each one), and press Connect. Use Sync now (the refresh icon on a connector) to pull in new content.

ConnectorWhat it importsWhat you need
Google DriveDocuments, presentations and spreadsheetsYour Google account, or a public folder link
DropboxFiles from your DropboxA Dropbox app key and secret
OneDrive / SharePointFiles from OneDrive and SharePoint sitesAn Azure app registration (client ID, tenant ID, secret)
NotionPages and databasesA Notion internal integration token
ConfluenceDocumentation and wiki pagesWiki URL plus API token (Cloud) or username/password (Server / Data Center)
SlackMessages, threads and files from channelsA Slack bot token
ZoteroYour research library — papers, annotations, notes and PDFsA Zotero API key and your user or group ID

YouTube is listed as Coming Soon.

Google Drive Google Drive ​

Google Drive can be connected two ways:

  • My Google Drive — sign in with Google and allow access. Afterwards, use Import from Drive on the connector to pick the files you want. You can also Back up collection to copy a collection's original files to your Drive.
  • Shared folder link — read a folder someone shared as "Anyone with the link". No sign-in is involved and nothing is written back; choose whether to include sub-folders and which collection to import into, then press Connect and then Sync now.

Dropbox Dropbox ​

  1. Open the Dropbox App Console and create an app with "Full Dropbox" or "App folder" access
  2. Add the redirect URI shown in the connector form
  3. Copy the App key and App secret into the form and press Connect

Notion Notion ​

  1. Open Notion Integrations and create a new internal integration
  2. Copy the Integration Token
  3. Share the Notion pages you want to sync with the integration
  4. Paste the token into the connector form and press Connect

OneDrive / SharePoint ​

  1. In the Azure portal, create an app registration and a client secret
  2. Add the Microsoft Graph permissions Files.Read.All and Sites.Read.All and grant admin consent
  3. Enter the client ID, tenant ID and secret (and optionally site URLs) in the connector form

Confluence ​

  • Cloud: create an API token at Atlassian and use your email as the username
  • Server / Data Center: use your username and password
  • Enter the wiki URL; optionally a space key or page ID to limit the sync

Slack ​

  1. Create (or open) an app at Slack API Apps
  2. Under OAuth & Permissions, add the bot scopes channels:history, channels:read, groups:history, groups:read, users:read
  3. Install it to your workspace and paste the Bot User OAuth Token

Zotero ​

  1. Create an API key with read access at Zotero API Keys
  2. Note your User ID (shown on the same page) — or use the Group ID for a group library
  3. Enter the key, library ID and library type in the connector form

Academic sources ​

Academic databases are not connectors — they are searched live:

  • In chat and Deep Research, academic search covers arXiv, PubMed and Semantic Scholar
  • Import papers creates documents from DOI, arXiv or PubMed identifiers, with metadata from Semantic Scholar — see Uploading Documents
  • Find open-access PDF looks up a free copy of a paper via Unpaywall

MCP integrations ​

Settings → Integrations connects remote MCP servers to your chat assistant. Enabled servers' tools are available in every conversation.

  • Recommended servers — a catalog of verified servers grouped by topic (humanities and primary sources, sacred texts, academic sources, search and the web, data and computation). Switch one on to enable it; some need an API key, and Get a key links to where you get one.
  • Your servers — Add server to connect any remote MCP server over HTTP or SSE, with an optional bearer token. Test connection checks that it responds and reports how many tools it offers.

Connect your agent ​

Scrapalot also works the other way round: it is an MCP server, so your documents, research and memory become tools in Claude Code, Claude Desktop, Gemini CLI, or any MCP client.

  1. Go to Settings → Integrations → Connect your agent
  2. Click Create key and name it after the client that will use it
  3. Copy the key — it is shown only once
  4. Choose your client and copy the ready-made configuration (it already contains the MCP endpoint and your key)
  5. Ask your agent to list your collections — it should see your library right away

Treat keys like passwords

An API key has full access to your account, not just these tools. You can disable or delete a key at any time from the same screen; any agent using it stops working immediately.

Discord ​

Link your Discord account in Settings → Account to get the "Scrapalot Verified" badge next to your name on the community Discord server.

Troubleshooting ​

Connectors are locked ​

Connectors are a Pro feature. On the free Researcher plan the Connectors tab shows an upgrade prompt.

Authorization failed ​

Check the credentials against the setup steps in the connector form. For Dropbox, the redirect URI in your Dropbox app must match the one shown in the form exactly. For Google Drive with a shared link, the folder must be shared as "Anyone with the link".

Nothing was imported ​

  1. Press Sync now on the connector (or Import from Drive for your own Google Drive) — connecting alone does not import files
  2. Check that the source actually grants access (Notion pages shared with the integration, Slack bot added to the channels, Microsoft Graph admin consent granted)
  3. Check the collection you imported into, and the Upload tab for processing errors

MCP server does not connect ​

Use Test connection. Make sure the URL is a reachable http(s) address and that the auth token is correct — some catalog servers refuse a token they do not expect, so leave the field empty for those.

Learn More ​

Open-core — Community Edition under AGPL-3.0 · Hosted product is proprietary.