What's new in Blocks

Release Notes

This release significantly strengthens Blocks Cloud’s data governance and access control capabilities, introducing a unified security model in Data Gateway that combines RBAC, ABAC, Row-Level Security (RLS), and Column-Level Security (CLS) with policy inheritance for enterprise-grade enforcement.

In addition to dynamic, identity-aware data filtering, the release introduces Data input validation using regex expressions, ensuring structured and compliant data entry at the schema level.

Further enhancements include:

• A new Localization endpoint for retrieving specific language keys efficiently
• Project ownership transfer for improved operational continuity
• Stability improvements and critical bug fixes

Together, these updates enhance data security, governance flexibility, and administrative control—while reducing application-level complexity and strengthening alignment with modern Zero Trust and multi-tenant architecture principles.

---

Data Gateway – Unified Data Security & Policy Engine

Data input validation

In the schema structure section click on the icon under the validation column to set up validation for that column (of the schema).

Users can add regex expressions to their columns. This means every time they enter a value, it must pass the regex expression, only then data entry will be made. There can be only one expression per column. If validation fails a custom message can be provided. The regex setup can be set to active or inactive.

Blocks Cloud Data Gateway now provides a comprehensive, rule-based security model that combines Role-Based Access Control (RBAC) and Attribute-Based Access Control (ABAC) to enforce fine-grained data governance at both the row and column levels.

This enables organizations to define dynamic, identity-aware policies that control not only who can access data, but also which records and which fields are visible per request.

Multi-Layered Access Control Model

1. Role-Based Access Control (RBAC)

Access to schemas, endpoints, and operations is governed by roles and permissions assigned to users. RBAC establishes baseline access boundaries at the resource level.

2. Attribute-Based Access Control (ABAC)

Beyond static roles, Data Gateway supports dynamic evaluation of access rules using:

• User identity attributes
• Token claims
• Organizational context
• Schema fields
• Static or derived values

This allows policies to be evaluated at runtime based on contextual conditions.

Row-Level Security (RLS)

Row-Level Security enables administrators to filter data records dynamically per user or role.

Policies are defined using a rule engine that supports operators such as:

• Equal / Not Equal
• In / Not In
• Starts With / Ends With
• Is Null / Is Not Null

Rules can reference:

• Schema fields (e.g., CreatedBy,OrganizationIds,Tags,Language)
• Authentication context (e.g., Auth.UserId,Auth.OrganizationIds)
• Static values

Rules can be combined using logical conditions (e.g., “all conditions must match”), allowing complex policy definitions.

Example Use Cases

• Users can only access records they created
• Multi-tenant isolation based on OrganizationIds
• Department-based filtering
• Soft-deletion enforcement (IsDeleted = false)
• Language-specific content visibility

RLS ensures that only authorized records are returned in a query, without requiring custom filtering logic in application code.

Column-Level Security (CLS)

In addition to row filtering, Data Gateway supports Column-Level Security, enabling granular control over field visibility within a schema.

CLS allows administrators to:

• Hide sensitive fields (e.g., financial data, internal flags, audit metadata)
• Expose specific columns based on role or identity attributes
• Apply rule-based conditions to determine field-level access
• Enforce dynamic field masking strategies

Column policies are evaluated at request time, ensuring that only authorized fields are included in the response payload.

Policy Inheritance & Layered Governance

Column-level policies can optionally inherit rules defined at the schema (row) level.

This inheritance model provides:

• Consistency between row and column enforcement
• Reduced duplication of policy definitions
• Simplified administration
• Flexible overrides when needed

Administrators may:

• Apply schema-level policies globally
• Extend them at the column level
• Override specific fields with stricter or alternative rules

This layered design supports scalable governance across complex data models.

Architectural Benefits

• Enforces least-privilege data access
• Eliminates hardcoded authorization logic in backend services
• Centralizes policy enforcement at the data layer
• Strengthens multi-tenant isolation
• Aligns with Zero Trust architecture principles
• Supports compliance-driven field-level restrictions (e.g., GDPR-style data minimization)

In Summary

Data Gateway now delivers a unified, enterprise-grade policy engine that combines RBAC and ABAC to enforce:

• Resource-level control
• Row-level filtering
• Column-level visibility restrictions
• Identity-aware runtime evaluation

This provides secure, scalable, and highly granular data access control—without requiring application-level customization.

---

Localization - New Endpoint

A new endpoint has been introduced to allow applications to retrieve specific localization keys without downloading the entire language file.

curl -X POST 'https://api.seliseblocks.com/uilm/v1/Key/GetsByKeyNames' \
--header 'accept: application/json' \
--header 'x-blocks-key: <your-x-blocks-key>' \
--header 'Authorization: Bearer <your-access-token>' \
--header 'Content-Type: application/json' \
--data '{
  "keyNames": [
    "ABOUT",
    "ACCEPTED",
    "ACTION"
  ],
  "moduleId": "<optional-module-id>",
  "projectKey": "<your-x-blocks-key>"
}'

Check out the swagger for more details.

You will need to add the necessary permissions for this request. Use api::key::getsbykeynames in the Resource field. See our guide on how to add permissions.

---

Blocks Cloud – Ownership Transfer

Blocks Cloud now supports project ownership transfer, allowing the current owner to securely assign ownership to another user within the project.

Once ownership is transferred, the new owner assumes full administrative privileges, including the ability to:

• Create and manage environments
• Add or remove users from the project
• Update and edit the project name
• Perform all actions previously reserved for the original owner

This feature ensures operational continuity, simplifies role transitions, and supports evolving team structures without requiring platform-level intervention.

---

Bug fixes

• Fixed Data Gateway instances were not reloading
• Fixed Create role endpoint was getting a 403 error if the user is not project owner

Release Notes

This release strengthens Blocks Cloud with secure Magic Links, OIDC-based SSO via Blocks IDP, enhanced Selise AI agent integration, improved localization change detection, AWS S3–compatible storage support, and a redesigned environment and user management experience.

Together, these updates improve automation, governance, flexibility, and AI-driven workflows across the SELISE Blocks ecosystem.

---

Utilities - Secure Redirects & Backend Action Triggers

Blocks Cloud now introduces Magic Links, enabling teams to generate secure, time-limited URLs for streamlined user flows and automated backend actions.

Magic Links are designed for flexibility and control, supporting configurable usage limits, expiration dates, authentication scopes, and environment-specific base URLs.

Two Types of Magic Links

Redirect Magic Links Generate shortened URLs that securely redirect users to a configured destination while maintaining authentication context. Ideal for passwordless login, invitations, and guided user journeys.

Action Magic Links Trigger predefined backend HTTP requests (GET or POST) when accessed. These links execute actions such as email verification, workflow triggers, or system events—without requiring user redirection.

Key Capabilities

• Configurable usage limits
• Optional automatic expiration
• Scoped access via client credentials
• Support for custom headers, payloads, and query parameters (Action type)
• Customizable short URL base configuration
• Centralized dashboard for viewing, filtering, and managing links

Magic Links provide a secure and extensible mechanism for handling authentication flows, event-driven automation, and controlled access patterns across environments.

Identity - OIDC

OpenID Connect (OIDC) Support for Blocks IDP

This release adds OpenID Connect (OIDC) support, enabling applications to use Blocks Identity Provider (IDP) as a standards-compliant Single Sign-On (SSO) solution.

Applications can now integrate with Blocks IDP using the Authorization Code flow, enabling secure, standards-compliant user authentication while delegating identity management to the Blocks platform. This allows any third-party application to use Blocks IDP as its SSO provider.

Blocks Construct also supports this capability—by enabling Bring Your Own SSO and implementing Blocks OIDC—allowing Construct-based deployments to participate in the same unified single sign-on experience across environments.

Key benefits include:

• Centralized authentication through Blocks IDP
• Secure, industry-standard OIDC authorization flow
• Seamless SSO availability across Blocks-based deployments
• Reduced need for custom authentication logic in applications

This enhancement strengthens identity integration within the SELISE Blocks ecosystem and provides a scalable foundation for SSO-enabled applications.

Localization

A new endpoint is now available to help determine whether localization content has changed, without requiring a full language file download. This allows applications to efficiently detect when an update is necessary by checking the publication history of language resources.

New endpoint: /Key/GetLanguageFileGenerationHistory Detailed specifications are available in the API documentation Swagger.

The endpoint returns versioned metadata for published language files, including creation time and project context, enabling lightweight change detection.

Newly Supported Languages

• Bengali (Sylheti)
• Bengali (Chittagonian)

This update improves localization efficiency while expanding language coverage for regional variants.

Selise AI – Blocks Cloud Agent Integration

Selise AI continues to evolve as a unified conversational interface.

Previously, users could interact with multiple large language models (LLMs) within a single GPT-style experience. With this release, Selise AI now supports direct interaction with agents configured on Blocks Cloud.

This enhancement allows users to connect to agents powered by the Blocks RAG architecture, leveraging:

• Dedicated knowledge bases
• Custom tunings and configurations
• Context-aware retrieval
• Enterprise-ready deployment controls

By integrating Blocks Cloud agents into Selise AI, organizations can move beyond general-purpose LLM conversations and interact with domain-specific, retrieval-augmented assistants tailored to their own data and business logic.

This update strengthens the bridge between conversational AI and the broader **SELISE Blocks ecosystem, enabling more intelligent, contextual, and controlled AI experiences.

UDS – AWS-Compatible Object Storage Support

UDS (formerly Data Gateway) now supports AWS S3–compatible object storage, enabling integration with alternatives such as MinIO.

This enhancement allows organizations to use self-hosted or third-party S3-compatible storage solutions while maintaining compatibility with existing AWS-style APIs. It provides greater flexibility in storage architecture, supports hybrid and on-prem deployments, and reduces dependency on a single cloud provider.

Blocks Cloud

A new dedicated section has been introduced for navigating Environments, Project Settings, and People Management, creating a more structured and intuitive user experience.

This update streamlines navigation across environments and improves visibility and control over user access. Teams can now:

• Invite users to multiple environments
• Clearly see who has access to each environment
• View and manage user profiles
• Manage project-level settings from a centralized location

This release focuses on isolation, identity, and intelligent communication. Each project and environment now has its own schema, ensuring complete data separation. The IAM module gains new authentication options, while the AI, Communication, and Managed Services sections see meaningful functional and usability updates.

---

Release Notes

v3.7.6 introduces dedicated Data Gateway instances, smarter identity management with client credential support, and context-aware AI translations. These enhancements improve security, flexibility, and collaboration across SELISE Blocks Cloud.

---

Projects & Environments

Every project and environment now has its own schema, replacing the previously shared model.
This provides complete isolation between environments, ensuring that project-specific data, rules, and configurations remain independent.

---

IAM (Identity & Access Management)

New Features

• External IdP JWKS Support — Administrators can now integrate third-party authentication providers such as Auth0, Keycloak, or any custom OpenID Connect (OIDC)–compliant service through JWKS (JSON Web Key Set) configuration. By sharing the provider’s JWKS endpoint within IAM settings, Blocks Cloud can directly verify tokens issued by these external systems. This enhancement allows organizations to bring their existing user base into Blocks Cloud without requiring re-registration or manual synchronization. Users authenticated through their existing IdP can seamlessly access Blocks Cloud resources while benefiting from native identity, authorization, and LMT (Logs, Metrics & Tracing) capabilities. In essence, teams can now unify their external authentication ecosystem with Blocks Cloud — maintaining centralized identity management externally while still gaining full observability and policy control within the platform.
• Client Credentials Grant — Added support for the Client Credentials OAuth 2.0 flow, enabling secure, machine-to-machine authentication. This allows backend services or applications to access APIs using client ID and secret without requiring user interaction.
• SSO Renaming — The “Social” tab has been renamed to SSO (Single Sign-On) for clarity and alignment with industry standards.

Improvements

• Tab Sequence Updated — The sequence of authentication tabs has been restructured for a more logical setup flow.

---

Data Gateway

New Features

• Preview Section — Added a powerful preview interface allowing users to view endpoints for GraphQL databases, auto-generate Create/Read/Update/Delete (CRUD) queries, and inspect schema structure in JSON format.
• Dedicated Instances per Project/Environment — Each project or environment now comes with its own Data Gateway instance, ensuring dedicated GraphQL resources and complete data separation.

---

AI

New Features

• Custom Agent Banner — Organizations can now add branded banners to their agents. The banner appears at the top of the agent interface and supports image formats, allowing teams to display announcements, promotions, or visual identity elements.

---

Communication

New Features

• Sinhala Language Added — Sinhala is now supported as part of the translation system, expanding accessibility.
• Key-Specific Context Provision — AI translation models often produce linguistically correct results but can miss contextual nuances — especially when the same phrase or label carries different meanings depending on where it appears in the interface. With key-specific context provision, developers can now attach contextual metadata to each translation key. This additional layer of information helps the AI understand where and how the text is used — whether it’s part of a button, tooltip, message, or title — allowing translations to stay accurate not just in language, but also in intent and placement. As a result, automatic translations across the platform become more reliable, natural, and domain-aware, minimizing errors that come from one-size-fits-all translations.

---

Managed Services

New Features

• Managed Services — The Managed Services capability allows users to connect their own backend services with the core features of SELISE Blocks Cloud — effectively extending the platform’s observability, identity, and data capabilities into their own systems. In this release, users can integrate Blocks’ native Logs, Metrics & Tracing (LMT) features directly into their managed services. This enables unified monitoring, debugging, and performance analysis across both internal and Blocks-hosted components. Looking ahead, this integration layer will expand beyond LMT to include additional platform services — allowing teams to view their service usage, operational health, and even vulnerability scans (via SCA, SAST, and DAST) within the same dashboard.
• Future Enhancements — In upcoming releases, users will also be able to view service health and scan for vulnerabilities using SCA (Software Composition Analysis), SAST (Static Application Security Testing), and DAST (Dynamic Application Security Testing) integrations — bringing proactive observability and security into one place.

---

Documentation

• User Docs Site Launched — The official documentation for SELISE Blocks Cloud is now live at docs.seliseblocks.com/cloud. Explore detailed guides, configuration steps, and API references.

---

Bug Fixes

• Fixed UI responsiveness issues in the “Export Keys” section.
• Updated “Reload Schema” API to dynamically load projects. (#1680)
• Fixed agent scrolling issues on mobile devices. (#1497)
• General responsiveness improvements across modules.
• Fixed language data migration bug where duplicate keys were being created.
• Resolved issue where Test Construct users were unable to log in due to a cache problem. (#1645)

---

Summary

SELISE Blocks Cloud v3.7.6 continues the journey toward isolation, intelligence, and integration — offering developers more control, businesses stronger identity options, and teams smarter tools for communication and observability.

This release introduces the Data Gateway service to Construct along with an early-stage integration of Storybook for component previews.

---

Release Notes

This release expands flexibility and intelligence across the platform. Users now have better control over AI Agents through customizable enhancements, user workflow tracking, and history and context retention with UILM. AI-powered Logs, Metrics & Tracing enhances environment migration starting with the UILM model. Together, these improvements make the platform more adaptable, resilient and effective for a wide range of applications.

---

Environment Migration

Our cloud platform now supports environment migration, giving users the flexibility to move data between environments. This capability minimizes workloads, simplifies testing, and reduces resource consumption when transferring projects. Support begins with the UILM language model. Migration ensures better traceability and logs for auditing. Migration support will gradually extend across additional models and services.

---

UILM

UILM now includes language history with recast capability, enabling users to track changes and restore previous versions when needed. This is especially useful for recent project introductions or errors. If you want to compare different states, or simply rewind to a prior task to enhance functionality, this major change enables safe and reversible UILM edits. Boosts safety and flexibility for your workflows.

---

Bug Fixes

• Resolved issue where API calls occurred with decoupled modules in "Create New Key".
• Fixed non-functional reset button appended with no previous data available.
• Corrected alignment issues within the module selection field.

---

Logs & Tracing

Our Logs, Metrics & Tracing (LMT) now has AI support, allowing users to query logs, metrics, and traces in natural language. You can even search directly by trace ID making it easier to diagnose issues and ensure integrity.

New Features

• Added support for searching by trace ID.
• Enhanced Ask AI, providing descriptive insights powered by AI.

---

AI Agents

AI Agents now give users more control through adjustable parameters, ensuring the assistant behaves as needed – whether that means more focused, more creative, or more efficient.

New Features

• Context window – set how many past messages are remembered.
• Context preservation – additional summarization and conversation tools to preserve context.
• Embeddings models:
  • Embedding v2 base: standard batch size for large models
  • Embedding v3 small
  • Embedding v3 large
• Chunking Strategy – choose how text is split into chunks:
  • Sentence
  • Semantic
  • Windowed
  • Custom
• LTM Conversation – added source links under each response to trace the origin of information.
• Dialog reset – history of the conversation page has been changed.
• The agent playground page has been redesigned.

---

Improvements

• Enhanced response quality with improved retrieval re-ranking, including semantic clustering for more accurate and relevant results.
• Changed the trace accounts with relevant focus.
• Introduced a more robust system based on the Conversation API instead of query-based interaction for better processing and retrieval.

This release brings significant improvements to UILM, new functionality in AI and platform modules, as well as multiple fixes for stability and usability.

---

New Features

UILM Timeline

• Introduced Get API for retrieving timeline data.
• Added Show Key Timeline view to highlight critical entries.
• Enabled Save Timeline capability for persisting updates.

AI

• Conversations from bots now display the source of information.

Project Settings

• Users can now remove repository access directly from project settings.
• Introduced search filters for faster repository navigation.

Deploy & Observe

• Repositories can now be deployed to custom domains.

---

Enhancements

• Refactored UILM frontend components with improved consistency and performance.

---

Fixes

---

UILM

• Resolved import/export issues.
• Fixed auto-translate behavior for more reliable localization.
• Addressed multiple frontend refactoring bugs.

General

• Improved overall frontend stability across modules.

---

This release brings significant improvements to UILM, new functionality in AI and platform modules, as well as multiple fixes for stability and usability.

---

New Features

UILM Timeline

• Introduced Get API for retrieving timeline data.
• Added Show Key Timeline view to highlight critical entries.
• Enabled Save Timeline capability for persisting updates.

AI

• Conversations from bots now display the source of information.

Project Settings

• Users can now remove repository access directly from project settings.
• Introduced search filters for faster repository navigation.

Deploy & Observe

• Repositories can now be deployed to custom domains.

---

Enhancements

• Refactored UILM frontend components with improved consistency and performance.

---

Fixes

---

UILM

• Resolved import/export issues.
• Fixed auto-translate behavior for more reliable localization.
• Addressed multiple frontend refactoring bugs.

General

• Improved overall frontend stability across modules.

---