Connectors

Overview

Lakecode Enterprise integrates with Slack and Jira so your workflows connect to the tools your team already uses. Connectors deliver evidence packs, approval requests, and debug summaries without requiring anyone to leave their existing tools.

Slack

Setup

Slack integration uses an incoming webhook. Create one in your Slack workspace and store it in Databricks Secrets:

# Store the webhook URL as a secret
databricks secrets put-secret lakecode slack-webhook \
  --string-value "https://hooks.slack.com/services/T.../B.../xxx"

Then configure the connector in your AssistantSpec:

connectors:
  slack:
    webhook_url_secret: lakecode/slack-webhook
    default_channel: "#data-ops"
    approval_channel: "#data-approvals"

What gets posted

• Evidence pack summaries — after a /prove workflow completes, a summary is posted with key findings and a link to the full evidence pack
• Approval requests — when a workflow hits a WRITE or DESTRUCTIVE step, an interactive message is posted with Approve / Reject buttons
• Debug reports — after a /debug workflow, root cause analysis and suggested fixes are posted
• Workflow failures — if a workflow errors out, a notification is posted with the error context

Channel routing

You can route different notification types to different channels:

connectors:
  slack:
    webhook_url_secret: lakecode/slack-webhook
    channels:
      evidence: "#data-quality"
      approvals: "#data-approvals"
      debug: "#data-oncall"
      errors: "#data-oncall"

Jira

Setup

Jira integration uses the Jira REST API with an API token. Store credentials in Databricks Secrets:

# Store Jira API token
databricks secrets put-secret lakecode jira-token \
  --string-value "your-jira-api-token"

# Store Jira user email (for API auth)
databricks secrets put-secret lakecode jira-email \
  --string-value "service-account@acme.com"

Then configure in AssistantSpec:

connectors:
  jira:
    base_url: https://acme.atlassian.net
    api_token_secret: lakecode/jira-token
    email_secret: lakecode/jira-email
    default_project: DATA
    evidence_label: lakecode-evidence

What gets created

• Evidence tickets — after a /prove workflow, findings can be attached to an existing ticket or a new ticket is created with the evidence summary
• Debug tickets — /debug workflows create or update an incident ticket with root cause analysis
• Data quality issues — when evidence packs surface data quality problems, a ticket is created with the details and suggested remediation

Ticket mapping

You can map workflow types to Jira issue types:

connectors:
  jira:
    base_url: https://acme.atlassian.net
    api_token_secret: lakecode/jira-token
    email_secret: lakecode/jira-email
    default_project: DATA
    issue_types:
      evidence: Task
      debug: Bug
      data_quality: Bug
    custom_fields:
      severity: customfield_10042
      data_source: customfield_10043