Changelog

Boltic Table AI - Manage Boltic Tables with AI

New feature

Boltic Table AI is now live - an intelligent assistant that lets you create, query, and manage your entire Boltic Tables setup using plain English. No SQL, APIs, or manual setup required. Powered by LangGraph ReAct and OpenAI GPT-5.4, it understands intent and executes actions instantly.
‍

What's New

• Boltic Table AI (LangGraph ReAct) – A fully conversational AI agent connected directly to your Boltic Tables
• Natural Language Operations – Describe what you want, and the agent figures out the how
• Streaming Responses – Get real-time feedback as the agent works through your request
• Context-Aware Conversations – Session-based chat history means the agent remembers what you've discussed
• Smart Intent Dispatcher – LLM-powered routing automatically directs your request to the right agent - Database or Workflow

‍

What Boltic Table AI Can Do

Area Supported Operations

• Databases - List, Create, Get, Update, Delete
• Tables - List, Get, Create, Update, Delete
• Records - List, Get, Create, Update, Delete
• Fields - List, Get, Create, Update, Delete
• Indexes - Get Index Types, List, Create, Delete
• Web - Fetch and read content directly from URLs
  ‍

How It Works

Open the Boltic Table AI from your Boltic dashboard

1. Type your request  - "Create a new table for customer orders" or "List all records where status is pending"
2. The Intent Dispatcher routes your message to the right agent automatically
3. Boltic Table AI interprets your intent and executes the operation using Boltic Tables API
4. Get a real-time streaming response as actions are completed
5. Continue the conversation - the agent retains full context throughout your session

‍

‍

Why It Matters

• Manage complex database operations without writing a single line of SQL or code
• Execute CRUD and schema operations in seconds through natural conversation
• GPT-5.4 powered intelligence understands nuanced, multi-step requests accurately
• Smart routing ensures the right agent always handles the right task
• Session history stored in PostgreSQL with optional Redis caching for speed
• Fully integrated within Boltic's governed, reliable infrastructure
• Drastically reduce time spent on database management and schema operations
  ‍

Intent Dispatcher - Smarter Routing, Every Time

• The built-in Intent Dispatcher uses LLM-based routing to understand the nature of your request and send it to the right place:
• Mentions of tables, databases, records, fields, schema, CRUD → Boltic Table AI
• No manual switching. No confusion. Just seamless, intelligent routing.
  ‍

Boltic Table AI transforms how your team interacts with data - making powerful database operations as simple as sending a message.

Boltic Workflow - Real Data Visibility in Execution Logs

Enchancement

Overview:

• Previously, input logs in Execution History and Run Logs displayed unresolved placeholders (e.g. {{variable}}), limiting visibility into the actual payload sent during workflow execution. This made debugging dependent on assumptions or external validation.
• With this release, input values are now resolved and surfaced at runtime.

What’s Changed:

• Input logs now display resolved runtime values instead of raw {{}} placeholders
• Dynamic variables are evaluated during execution and persisted in logs
• Logs reflect the exact payload sent to Temporal for each workflow run

Why This Matters:

• Deterministic Debugging - Inspect actual runtime inputs without reconstructing variable states
• End-to-End Observability - Trace input data across execution stages with full fidelity
• Faster Root Cause Analysis - Identify null, malformed, or unexpected values instantly
• Improved Execution Confidence - Validate data integrity across workflow runs
  ‍

Impact:

• Eliminates ambiguity in input inspection
• Reduces reliance on external debugging and log correlation
• Accelerates issue resolution and improves workflow reliability
  ‍

Applicability:‍

Applies to all workflows leveraging dynamic variables in Execution and Run Logs

New Integration: Salla - Triggers & Activities Now Live

New feature

We’re thrilled to announce that Boltic now officially supports Salla - one of the most powerful and widely adopted e-commerce platforms in the Middle East. You can now connect your Salla store directly within Boltic and build powerful, real-time automations - no custom setup required.
‍

What's New:

• Salla as a Trigger – Automatically start workflows when key store events occur

• Salla as an Activity – Perform actions on your store within workflows

• Wide event coverage – Orders, customers, inventory, products, and more
‍

How It Works:

1. Connect your Salla store from the Integrations panel

2. Choose Salla as a Trigger to start workflows (e.g., new order, customer update)

2a. Or use Salla as an Activity to push data or take action mid-workflow

2b. Combine Salla with other integrations to build end-to-end automations
‍

Why It Matters:

• Native support for Salla’s full commerce event ecosystem

• Automate repetitive store operations - no code needed

• Seamlessly connect with CRMs, ERPs, and notification tools

• Scale workflows with reliability and governance built in

• Tailored for Middle Eastern commerce use cases
‍

Boltic’s Salla integration brings your store data to life:

Automate faster, respond instantly, and scale your commerce workflows with ease.

Boltic Workflow - Activate only one Version at a time

Enhancement

We've updated how workflow versions are managed in Boltic.

What's changed:

• When you publish a new version of a workflow, all previous versions will automatically be marked as Inactive
• Only 1 version of the respective workflow will be UP and RUNNING
  ‍

Why This Matters:

• No conflicts - Eliminates the risk of multiple versions running simultaneously and causing unexpected behavior
• Predictable execution - Always know exactly which version of your workflow is live
• Easier debugging - Troubleshoot with confidence knowing only one version is active at a time
• Cleaner version control - Keeps your workflow history organised and easy to track
• Faster rollouts - Publishing a new version instantly takes over without any manual deactivation steps
• Cost effective - Boltic's metering system bills only for active usage - no charges for idle or redundant versions

Impact:‍

Clients get a more stable, conflict-free workflow experience with tighter cost control - only the version that matters stays active.

Applicability:‍

These changes apply to all workflows created after 1st Jan 2026.  All older workflows will have older versions running too.
‍

New Integration: Google Tasks - Activity Support Now Live‍

New feature

Boltic now supports Google Tasks, letting you connect your account and seamlessly create, update, and manage tasks within your workflows - no custom setup required.

‍What's New:

• Google Tasks as an Activity – Perform actions on your Google Tasks account mid-workflow
• Full task management support – Create tasks, update task details, mark completions, and manage task lists programmatically
• Seamless workflow integration – Use Google Tasks alongside any trigger to assign and track work at the right moment
  ‍

How It Works:

1. Connect your Google Tasks account from the Integrations panel
2.  Add Google Tasks as an Activity step within any workflow
3. Configure the action - create a task, update an existing one, or manage task lists
4.  Combine with other integrations to build end-to-end task and productivity pipelines
   ‍

Why It Matters:

• Native support for Google Tasks' full task management ecosystem
• Automate task creation and updates without writing a single line of code
• Connect task actions with CRMs, project management tools, and notification systems
• Keep your team's to-dos in sync across every workflow, automatically
• Tailored for teams already operating within the Google Workspace ecosystem

Boltic's Google Tasks integration brings your productivity workflows to life - assign the right tasks at the right time, automatically.

Introducing Direct PostgreSQL Access for Boltic Databases through Connection string

New Feature

Introducing Direct Boltic Database Access - connect to your Boltic-managed databases using a standard Postgres connection string.
‍

Key Features:

1. Standard Postgres Connectivity

• Connect Boltic Database using a Postgres connection string with any ORM, database client, or BI tool - no custom setup required.

2. Managed Infrastructure with Transparent Access

• All traffic flows through a secure architecture:
• Client → Gatekeeper → PgBouncer → Cloud SQL
• Ensuring performance, safety, and isolation.

3. Quota Enforcement

• Platform-level limits are enforced automatically, including write blocking when usage exceeds defined quotas.

4. Connection Pooling at Scale

PgBouncer manages and optimises database connections for high concurrency and efficiency.

5. Secure Access Control

• Connection strings visible only to database creators
• Access scoped per account and database
• Credentials can be reset anytime
• All tables and schema changes are tracked centrally to ensure consistency and governance.

6. User Access & Roles‍

Access to Boltic Databases is governed through predefined roles:

• Admin - Full database privileges (schema and data operations)
• Executive - Read/write access to tables and records
• Viewer - Read-only access to tables and records

This ensures controlled collaboration with clear separation of responsibilities.
‍

‍How It Works:

1. Create a database from the Boltic dashboard
2. Copy the Boltic database connection string
3. Connect using your preferred tool (PSQL, ORM, BI tool, etc.)
   ‍

‍Direct Boltic Database Access brings the power of Postgres into Boltic, so you can work with your data directly while still getting the platform’s control, security, and reliability. This means you can build and ship custom extensions/usecases for Fynd Commerce, Shopify, Wix, Magento, and more without worrying about setting up or maintaining infrastructure.

Multi-DB, Encrypted Columns and Indexing - Now Live!

Enhancement

Boltic Database (Boltic Tables) has been upgraded to deliver better scalability, security, performance, and developer experience. This release introduces new capabilities across database management, data protection, and query optimization to help you build and scale more efficiently.

What’s New

1. Multiple Databases per Account

• Create multiple databases within a single account
• Manage databases directly from the Boltic Console
• Easily separate environments (dev / staging / prod)
• Better isolation for different applications
• Scale cleanly as your data footprint grows

2. Encrypted Column Type - Built-In Data Security

We’ve introduced a new encrypted column type to securely store sensitive data directly in your tables.

Key highlights:

• Encryption is handled automatically by Boltic
• Ideal for PII and confidential information
• Optional deterministic encryption support same input produces the same encrypted output when enabled

3. Index Creation - Boost Query Performance

You can now create indexes on table columns directly in Boltic Tables.

This allows you to:

• Significantly improve query performance
• Clearly view and manage existing indexes per table
• Create and manage indexes directly from the table UI
• Support for single-column and multi-column indexes
• Choose the right index type based on your query pattern:
  
  • B-Tree: equality and range queries
  • Hash: exact match lookups
  • GIN: arrays and search use cases
  • SP-GiST: specialized and hierarchical data
  • BRIN: large, naturally ordered tables

‍
4. Revamped Tables UI

• Cleaner and more scannable list view
• Rich detail popup with contextual insights
• Side-panel record viewer for seamless navigation
• Inspect data without losing context

5. Table Templates

• Pre-built schemas for common use cases
• Comes with predefined columns and relationships
• Reduces setup time significantly
• Helps users get started faster

6. AI-Powered Table Filters

• Generate filters using plain English queries
• No need to remember column names or syntax
• AI suggests accurate query logic automatically
• Users can rate responses to improve accuracy

7. Usage Metering & Subscription-Based Blocking

We’ve added usage metering and enforcement based on your account’s subscription plan.

This includes:

• Accurate usage tracking at the account level
• Enforcement based on your subscription plan
• Automatic blocking when limits are exceeded

‍

Why this matters:

• Scale across multiple databases
• Secure sensitive data with encryption
• Optimize performance using indexes
• Stay in control of usage and costs

SQL Workspace for Boltic Database & Storage SDK for Boltic Storage

New Feature

We’re excited to introduce major updates across Boltic, enhancing Boltic Database with a powerful SQL workspace and extending Boltic Storage with a developer-friendly SDK.

‍

What’s New

Boltic Database - SQL Workspace

A complete SQL workspace with AI-powered queries, schema visualization, and built-in exports.

‍

1. SQL Editor with AI - Write & Run Queries Effortlessly

• Full-featured SQL editor with multi-tab support
• Run queries with instant results
• Format queries using Beautify
• AI assistance to generate and fix queries
• Smart error messages with actionable suggestions
• View results in Table/JSON  formats
• Export query results as CSV or JSON

2. Smart Queries - AI-Powered Query Suggestions

• Smart AI analyzes your database and generates relevant queries automatically
• Need more ideas? AI adapts to your tables and lets you generate additional queries in one click

3. Saved Queries - Reuse & Collaborate

• Save frequently used queries
• Quickly access and re-run saved queries
• Share queries across your team

4. Execution History - Full Query Audit Trail

• Track every query executed with complete visibility

5. Schema Visualizer - Understand Your Database Instantly

• Visual representation of your database schema
• View tables and their columns in a structured layout
• Quickly understand structure

‍

Boltic Storage - Storage SDK

We’re introducing a new SDK for Boltic Storage, enabling developers to seamlessly store, manage, and access files programmatically.

• Upload files to storage bucket
• Create and organize folders with ease
• Manage files efficiently at scale
• Control file visibility:
  
  • Public access
  • Private access
• Retrieve files programmatically

‍

We have also published Boltic Storage SDK

NPM: https://www.npmjs.com/package/@boltic/sdk

Doc: https://docs.boltic.io/sdk/storage/overview

‍

Why This Matters

• Run advanced SQL directly inside Boltic Database
• Reduce dependency on external database tools
• Improve team collaboration with shared queries
• Gain full visibility into query activity and exports
• Visualize your schema alongside querying

Combine AI + SQL + Visualization in one unified experience

Boltic Workflows – Failure Handling (Continue on Failure)

Enhancement

We’re excited to introduce failure handling controls in Boltic Workflows, designed to help you build more resilient and reliable automations.
This update gives you finer control over how your workflows behave when something goes wrong.

What’s New: Continue on Failure

You can now configure workflows to handle node failures gracefully instead of stopping execution entirely.

• New setting: Continue on failure (available in Advanced Options)
• Default: OFF
• When enabled, workflows can continue execution even if a node fails
• Unlocks alternative execution paths for better fault tolerance

‍

How It Works

1. When Enabled

• A failure handle appears on the node
• You can connect a separate path to this failure handle
• If the node fails, execution continues via this new failure path
• Define fallback logic, recovery steps, or alternate flows

2. When Disabled

• Any node failure will stop the entire workflow execution

3. Failure Context

The failure path receives context from the failed node, including:

• error → accessible using activity_name.error

This allows you to:

• Log failures
• Trigger alerts
• Build conditional recovery logic based on failure details

Why This Matters

• Build fault-tolerant workflows that don’t break on minor issues
• Add fallback paths and alternate flows
• Improve production reliability
• Gain greater control over error handling and execution flow

‍

No-Code API Creation with HTTP Request & Response Nodes

New Feature

Fynd Boltic now lets you build real-time, fully functional APIs using visual workflows — no backend coding required. With the powerful combination of HTTP Request and Response nodes, anyone can create custom APIs and webhook-based automations directly within Fynd Boltic.

HTTP Request Node – Start Your Workflow via HTTP

• Triggers workflow execution on receiving an HTTP request.
• Supports multiple HTTP methods: GET, POST, PUT, DELETE, etc.
• Captures path parameters, query strings, headers, and request body.
• Enables synchronous execution — waits for the workflow to complete before sending a response.
• Perfect for API endpoints, webhook handlers, and third-party integrations.

Response Node – Send a Custom Response

• Allows customizing the HTTP response sent after the workflow completes
• Supports setting status codes like 200 OK, 400 Bad Request, 500 Server Error
• Enables adding custom headers such as Content-Type, Authorization, and others
• Supports returning responses in multiple formats including JSON, HTML, Text, XML, YAML, CSV, PDF, Image, Binary, and more
• Provides full control over how the API communicates with the caller

This feature empowers both developers and non-technical users to build, test, and deploy logic-driven APIs visually - making it easier than ever to automate workflows, power integrations, or expose custom services in a secure and structured way.

Execution History Revamp

Enhancement

We’ve given the Execution History page a fresh look and powerful new features to improve visibility, traceability, and debugging:

• Execution Insights Panel - Instantly get a summary of your executions - view total runs, success/failure breakdowns, pending/running states, and average execution time all in one glance.
• Search & Filter Capabilities - Filter by Execution ID, status, date range, or trigger type to find exactly what you’re looking for - faster and easier.
• Detailed Execution Log - Track individual executions with enhanced granularity. Each row now includes status badges, execution timestamps, unique execution IDs, trigger sources, and exact time taken.
• Visual Feedback for Statuses - Quickly identify successful and failed executions using color-coded tags and intuitive icons.
• Real-time Refresh - Use the refresh button to instantly fetch the latest execution data without needing to reload the entire page.
  

Database Activities – Now Faster & More Powerful

Enhancement

We’ve upgraded database activities to deliver faster performance and broader compatibility with multiple databases - making your workflows smoother and more powerful than ever.

• Lightning-Fast Execution – Query execution time has improved from ~40 seconds to under 3 seconds, dramatically speeding up database interactions.
• Expanded Database Support – Now supports MongoDB, MySQL, PostgreSQL, MSSQL, BigQuery, and ClickHouse, giving users more flexibility with their data sources.
• Full CRUD Functionality – Perform INSERT, SELECT, UPDATE, and DELETE operations directly within your workflows.
• Beta Limitation on SELECT Queries – Currently supports up to 1,000 rows per query during beta phase.
• Seamless Integration – No need for external tools - manage and automate your data processes right within the platform.
• Explore all supported database apps – Database Apps

Faster database actions with full CRUD support and multi-database compatibility."

Test & Save button in Activity Nodes

Enhancement

You can now test your activity node configuration before saving it. This helps you instantly catch errors and ensures data flows correctly through your workflow.

• View real-time output before saving the node.
• Easily catch and fix misconfigurations early.
• Improves accuracy and confidence while building.
• You can also skip testing if you prefer
  

Dynamic Suggestions Panel & Data Pills

Enhancement

Typing {{ }} manually is now a thing of the past. A dynamic suggestions panel shows available data from previous nodes, making workflows faster and easier to build.

• Auto-suggests available data fields from earlier steps.
• Click to insert values - no need to remember field names.
• Instantly test and run nodes directly from the panel.
• Helps build and debug without navigating away from the form.

JSON-GUI Interface

New Feature

You can now create structured JSON using a simple, no-code visual builder. It’s designed to make your work faster and easier.

• Build JSON visually - no coding needed.
• Supports up to 5 levels of nested objects.
• Great for non-developers and beginners.
• Easily switch between the GUI and raw JSON editor anytime.