Skip to main content

Integrations

Data integrations keep your project’s document store in sync with the external systems where knowledge already lives. Each integration maps to a specific provider, stores files in the folder you choose, and tracks source metadata so indexes can refresh without manual uploads.

What Every Integration Provides​

  • Structured imports: Documents stream into your project using the same folder hierarchy you maintain in the source system.
  • Metadata fidelity: IDs, reference URLs, timestamps, and other source attributes are persisted to document metadata, enabling downstream filtering.
  • Safe re-runs: Loader run IDs and version checks prevent duplicate content and let incremental jobs remove or update changed records.

Built-in Connectors​

BotDojo includes production-ready connectors for common content systems:

Google Drive​

  • Exports Google Docs, Sheets, Slides, PDFs, text files, and more into BotDojo-friendly formats such as .docx, .md, or .pdf.
  • Rebuilds the Drive folder structure and preserves created / modified timestamps.
  • Maintains sharing URLs via meta.reference_url so flows can cite the source document.

Microsoft OneDrive & SharePoint​

  • Crawls the selected drive or folder, including nested directories.
  • Uses incremental change tracking so scheduled jobs only fetch new or modified files.
  • Mirrors owner, modification time, and other Microsoft metadata in the resulting documents.

Box​

  • Syncs any Box folder (or entire drives) while respecting your existing directory layout.
  • Leverages Box events for incremental updates, reducing API load on large libraries.
  • Copies Box metadata (collaborators, shared links, timestamps) into document metadata fields.

Zendesk Help Center​

  • Pulls help-center articles for every available locale.
  • Converts HTML articles to Markdown, bundles the source metadata, and stores the canonical reference_url for citations.
  • Organizes saved files into locale-specific folders within your project.

Salesforce​

  • Runs a user-defined SOQL query and writes one document per record.
  • Optional pruning removes records that disappear from the latest query results, keeping the folder in sync with Salesforce.
  • Retains record IDs, timestamps, and selected fields for metadata filtering.

Additional Connectors​

Other loaders follow the same lifecycle, including Google Shared Drives, Amazon S3 buckets, Airtable bases, Gong call libraries, Rocketlane workspaces, and more. Check the in-app integration catalog for the latest list or to enable beta connectors.

Monitoring Integration Runs​

Each integration records duration, document counts, and errors in the Usage tab. Use that history to validate schedules, inspect changes, and rerun jobs after resolving upstream issues.

With integrations in place, your indexes always draw from the freshest version of your knowledge base while keeping provenance intact.