Sources
Thirty of the two hundred and forty, in full.
Every source says what Sable reads from it. Read only, always, and nothing leaves your cloud.
Snowflake
Warehouse
Schemas, column names, query history and the tables people actually use.
BigQuery
Warehouse
Same reads, with your row-level policies applied to every answer.
Redshift
Warehouse
Schemas, sort and dist keys, and the joins that were never declared.
Databricks
Warehouse
Unity Catalog tables, lineage and column comments.
Postgres
Database
Schemas, foreign keys and views, read only.
MySQL
Database
Schemas and indexes, read only.
MongoDB
Database
Collection shapes inferred from a sample, not from a schema file.
dbt
Modelling
Model names, descriptions and tests, so a metric keeps its definition.
Looker
BI
Explores and measures, so an answer uses the same definition as the dashboard.
Tableau
BI
Published data sources and calculated fields.
Metabase
BI
Saved questions and the collections they sit in.
Salesforce
CRM
Accounts, opportunities and custom objects with their labels.
HubSpot
CRM
Companies, deals and the properties your team actually fills in.
Pipedrive
CRM
Deals, stages and the reasons attached to a loss.
Zendesk
Support
Tickets, tags, satisfaction scores and macro usage.
Intercom
Support
Conversations and the tags that survive a handover.
Front
Support
Shared inbox threads and who answered them.
Notion
Docs
Pages with their headings kept, so a sentence keeps its section.
Confluence
Docs
Spaces, pages and the runbooks nobody reads until they must.
Google Drive
Docs
Docs and Sheets, with the sheet a metric was invented in.
Slack
Conversation
Public channels only, and only where a bot is invited.
Gong
Conversation
Call summaries and the moment a customer said the word churn.
Segment
Events
Tracking plan, event names and the properties on them.
Amplitude
Events
Event definitions and cohorts, so a funnel keeps its meaning.
Stripe
Billing
Subscriptions, invoices and the reason a payment failed.
Chargebee
Billing
Plans, credits and the history of a price change.
Jira
Delivery
Issues, epics and the labels engineering uses in practice.
Linear
Delivery
Issues and cycles, with the project a change belonged to.
GitHub
Delivery
Pull requests and the files a metric definition lives in.
Workday
People
Headcount and team structure, used only for who can see what.