Changelog
All notable changes to dbt-Workbench are documented in this file.
[Unreleased]
Added
- Comprehensive SEO optimization across all documentation pages
- 7 new comprehensive guides:
- Migrate from dbt Cloud to dbt-Workbench
- dbt Testing Guide
- dbt Performance Optimization
- dbt Security & Compliance
- dbt Best Practices
- CI/CD Pipeline for dbt
- dbt Troubleshooting Guide
- dbt Documentation Guide
- Enhanced structured data with WebPage and Article schemas
- Improved meta tags for better search visibility
- Keywords frontmatter for all documentation pages
Improved
- Optimized titles and descriptions for SEO
- Enhanced Organization schema with social links
- Updated README with SEO keywords section
- Better internal linking between related pages
[1.0.0] - 2024-02-15
Added - Phase 12: AI Copilot
- Global AI assistant panel with workspace-aware conversations
- Runtime provider switching between direct LLM APIs and MCP servers
- SQL copilot actions in SQL Workspace (explain/generate/optimize/fix prompts)
- Lineage Q&A and run troubleshooting context actions
- Admin-configurable AI defaults, secrets, and MCP server registry
- Proposal-based execution flow for SQL/dbt actions with explicit confirmation
- Full AI prompt/response/tool/action audit persistence per workspace
Added - Phase 11: Git-Integrated dbt Workspace
- Workspace-scoped Git connections with branch switching, pull, push, and commit workflows
- In-app file tree with SQL/Jinja editor for models and YAML editor support for dbt configs
- Git-aware commit diffing, status, and history panels plus audit log visibility
- Role-aware editing controls for protected configuration files and conflict handling cues
- Model diff viewer
- Historical lineage browser
Added - Phase 10: Plugin Ecosystem
- Backend plugin manager with manifest validation, capability/permission checks, and lifecycle events
- Hot-reloadable plugins discovered from the configurable
PLUGINS_DIRECTORY(default./plugins) - Admin APIs to list, enable, disable, and reload plugins without restarting the server
- Workspace-scoped plugin configuration API for per-workspace settings
- Frontend marketplace and installed-plugins views with dynamic enable/disable controls
- Standardized plugin layout (
/plugins/<name>/manifest.json,backend/,frontend/,static/) - Safe opt-out via
PLUGIN_SYSTEM_ENABLED=falsefor minimal installations
Added - Phase 9: RBAC & Multi-Project
- JWT-based authentication (optional)
- Role-based access control (Viewer, Developer, Admin)
- Multiple workspaces with independent data
- Workspace switching and per-user defaults
Added - Phase 8: Data Catalog Layer
- Global fuzzy/prefix search across models, sources, exposures, macros, tests, tags, and columns
- Rich entity detail pages with dbt metadata, owners, tags, documentation, lineage previews, and column statistics
- Test health indicators surfaced in search, detail pages, and validation reports
- Source freshness visibility (max loaded timestamp, age, thresholds, status, last check)
- Persistent metadata enrichment for owners/tags/descriptions with optional edit controls
- Column-level descriptions, data types, nullability, and statistics synced from
catalog.json - Validation of missing documentation, owners/tags, failing tests, freshness gaps, and stale sources
Added - Phase 7: SQL Workspace
- SQL editor with syntax highlighting
- Dual-pane dbt model view showing editable source alongside read-only compiled SQL
- Query execution against configured database with dbt model runs using compiled SQL only, plus clear compile reminders when compiled artifacts are missing
- Environment-aware compilation and execution with role-based run restrictions
- Result profiling and statistics shared across custom SQL and dbt model runs
- Query history with execution mode, model references, and compiled SQL checksums
Added - Phase 6: Scheduler
- Cron-style scheduled runs with timezone support
- Email, Slack, and webhook notifications
- Environment-specific configurations
- Retry policies with exponential backoff
- Catch-up and overlap policies
- Historical run diagnostics with failure reasons, attempt timelines, and log/artifact links directly in the Schedules page
- Manual and cron-triggered runs expose queued/in-progress status with log links immediately after dispatch so "Run now" is easy to trace without waiting for completion
- Workspace-aware execution paths so scheduled dbt commands use the connected Git repository instead of the default project folder
Added - Phase 5: Advanced Lineage
- Column-level lineage derived from manifest and catalog artifacts
- Grouping by schema, resource type, and tags with collapsible aggregates
- Interactive expand/collapse of subgraphs to simplify large projects
- Upstream and downstream impact highlighting at model and column granularity
- Configurable defaults for grouping mode, graph depth, and column-level loading
Added - Phase 4: Metadata Persistence Layer
- PostgreSQL backend
- Historical model snapshots
Added - Phase 3: dbt Execution Engine
- Run dbt commands from UI
- Quick-launch Run/Test/Seed/Docs actions directly from the dbt Execution page via dedicated execution buttons (no separate command selector)
- Real-time log streaming
- Persist artifacts per run
- Dashboard surfaces the latest run status from execution history for consistency with the Run History page
- Seed-aware execution guardrails warn when seeds are required before running downstream models
Added - Phase 2: Live Metadata Updates
- Auto-detect changes to dbt artifacts
- Background watcher reloads metadata
- Frontend shows update indicators
- In-memory versioning
Added - Phase 1: Artifact Viewer
- Browse models, sources, tests
- Model details (columns, metadata)
- Basic lineage graph
- Runs list + statuses
- Dashboard overview
Categories
Changes are categorized as:
- Added: New features
- Changed: Changes to existing functionality
- Deprecated: Soon-to-be removed features
- Removed: Now removed features
- Fixed: Bug fixes
- Security: Security improvements