SaaS and enterprise apps
Salesforce, HubSpot, Workday, SAP, Dynamics, Shopify, Jira
Integration pipeline
Set up → pull → data store → push → review
New tenant integration
salesforce-prod
Connector: Salesforce · OAuth2 · active
Credentials: keystore · token refresh: on
Test connection: OK
hubspot-marketing
Connector: HubSpot · API key · active
Test connection: OK
Connect your stack
“Hundreds of systems, one way to connect them—workflows, APIs, and AI all use the same controlled setup.”
Platform
Connect CRMs, ERPs, cloud storage, databases, and custom APIs once—then run them from workflows, the REST API, and AI where you allow it. Not one-off automation tools sitting beside your real work.
Adoption context
Related workflow: Failed sync detection and retry · Adoption path
A connector is a ready-made bridge. An integration is your live connection on the platform—set up once, used everywhere.
Most platforms force a tradeoff: point-to-point scripts that break, automation connectors disconnected from your processes, or custom code that is expensive to secure and maintain.
Tealfabric sits in the middle—a broad connector library wired directly into workflows, your data, and access controls. Connections are not side projects; they are part of how work gets done.
How teams connect systems to processes and data.
SFTP + email
→ DataPool
Pull vendor data from email and SFTP, validate in a workflow, write to a Datapool.
Salesforce
nightly sync
A nightly workflow pushes and pulls account updates through Salesforce—with a clear history.
generic HTTP
legacy ERP
Legacy ERP via generic HTTP—credentials in the platform, not scattered scripts.
describe + execute
AI opt-in
AI describes what a connection can do, runs allowed actions, and leaves a reviewable history.
OAuth refresh
accounting API
OAuth accounting APIs with automatic token refresh for long-running automations.
Set up, pull from Salesforce, save to a data store, push to HubSpot—with a full history.
Set up connections once on one platform. A workflow pulls account data from an external system, writes enriched records to your data store, pushes updates through a second connection, and leaves a full trail you can review.
Integration pipeline
Set up → pull → data store → push → review
New tenant integration
salesforce-prod
Connector: Salesforce · OAuth2 · active
Credentials: keystore · token refresh: on
Test connection: OK
hubspot-marketing
Connector: HubSpot · API key · active
Test connection: OK
Breadth, reuse, reliability, and control—not connector count alone.
Documented actions, test before go-live, and a history of each run. Many connectors build trust; controlled execution is what matters.
CRM · ERP · cloud
messaging · files
A broad library across CRM, ERP, cloud, messaging, data, files, and protocols.
Business apps, cloud platforms, email and chat, databases, warehouses, CSV/Excel/PDF, REST, SOAP, GraphQL, webhooks, and AI services—each documents what it can send, receive, sync, and test before you go live.
test · activate
OAuth refresh
Your connections with secure credentials—not stored in workflow code.
Name, connect, test, and activate. OAuth apps stay connected with automatic token refresh. One Salesforce or SFTP connection can power dozens of workflows, API calls, and AI actions.
sync · queue
retry · cancel
Immediate runs or queued jobs with status, cancel, retry, and callbacks.
From quick lookups to overnight syncs—same connection, your choice of timing. Control in-flight work without a separate job runner.
workflow step
structured I/O
Call any configured connection from a workflow step with structured inputs and outputs.
Same execution path in the visual editor or custom code. No API keys in snippets. No shadow connections that live only in one script.
describe + execute
opt-in per integration
Discover and run with guardrails—turn on per connection.
AI can describe what a connection supports before calling it. AI access is off by default; you enable it where support automation and data enrichment make sense.
generic REST
internal APIs
Reach internal APIs, niche vendors, and legacy endpoints.
Named connectors for the majors; configurable method, path, headers, and body for everything else—within platform safety limits.
virtual schema
query + govern
Feed Datapools and shared schemas from connected systems.
Connections are not only request/response pipes. Query, transform, and govern external data alongside your operational datasets.
run history
rate limits
Test results, run history, queue visibility, rate limits, and review.
When something fails at 2 a.m., teams see what ran, for whom, and why—in one place alongside workflow and security activity.
Salesforce, HubSpot, Workday, SAP, Dynamics, Shopify, Jira
SMTP, Gmail, Slack, Teams, Twilio, WhatsApp Business
CSV/Excel/PDF, S3, SFTP, platform storage
MySQL, PostgreSQL, MongoDB, Redshift, BigQuery
AWS, Azure, GCP, webhooks, Kafka, MQTT
REST, SOAP, GraphQL
OpenAI, Anthropic, Gemini, speech APIs
Generic REST to any reachable HTTP API
Connect your stack to your processes
We walk through connector setup, immediate and queued runs, Datapool linkage, run history, and AI access you turn on per connection—on the same platform as your rules and operational data.
Pilots start with one live workflow—system connections, steps, and full history.