> For the complete documentation index, see [llms.txt](https://docs.ovaledge.com/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.ovaledge.com/release8.2/patch-releases/release8.2.1.md).
# Release8.2.1
This release introduces key enhancements and bug fixes across Chrome Extension, Data Catalog, askEdgi, and Connectors, improving authentication reliability, access control, LLM flexibility, and connector profiling and authentication.
**Key Highlights**
* **Browser Extension**: Resolved authentication failures after successful login, ensuring the extension completes the sign-in process correctly.
* **Data Catalog**: Added role-based access controls for code and query details, allowing users to access metadata while restricting sensitive code and query content across supported catalog views, APIs, and exports.
* **Integration**: Added Amazon Bedrock as an AI provider through secure AWS Cross-Account IAM AssumeRole integration, enabling supported AI capabilities to use Bedrock-based Claude models.
* **AI Capabilities**: Supports askEdgi analysis, Recipe creation, AI-generated descriptions, Term recommendations, and Data Classification Recommendations using supported Bedrock chat models.
* **Connectors**: Improved QuickBase profiling by removing dependency on the legacy XML API and added User Token authentication, improving profiling performance and expanding authentication options while retaining existing Username and Password support.
**Release Details:**
| Release Type | Release Version | Build<Release. Build Number. Release Stamp> | Build Date |
|---|
| Patch Release | Release8.2.1 | release8.2.1.821.9b63e21 | August 20, 2026 |
## Browser Extension
### Fixed
**Resolved Browser Extension Authentication Issue**
In the Browser Extension, an issue where authentication failed after a successful login, preventing the extension from completing the sign-in process, has been resolved.
The extension now correctly completes authentication after a successful login.
## Data Catalog
### New & Improved
**Control Access to Code and Query Details**
Users who could access a Code object could also view its full query or code, even when they only needed to browse or understand the object. This could expose sensitive business logic to users who did not need access to the underlying code.
With this enhancement, administrators can control access to code and query details through Application Security and assign access to specific roles. Users without permission can still access relevant metadata, while code or query details remain hidden across supported catalog views, APIs, and exported files.
{% hint style="warning" %}
As part of the migration validation from release **7.2.5.24 to 8.2.1**, the Data Catalog **Codes** tab was identified for an additional compatibility update to ensure uninterrupted access to code entries after migration. The update is included in the patch release (**Release8.2.1.1**), ensuring users can successfully view code data in the Data Catalog.
{% endhint %}
## Integration
### New & Improved
**Amazon Bedrock Integration**
OvalEdge now supports Amazon Bedrock as an AI provider for various AI capabilities across the platform.
Amazon Bedrock supports the following model configurations:
* **Embedding Models** – OvalEdge already supports Bedrock embedding models. This support continues in the current release. Embedding models are used to generate embeddings.\
The supported embedding models are:
* amazon.titan-embed-text-v2 (1024)
* amazon.titan-embed-text-v2 (512)
* amazon.titan-embed-text-v2 (256)
* amazon.titan-embed-text-v1 (1536)
* cohere.embed-english-v3 (1024)
* cohere.embed-multilingual-v3 (1024)
* **Chat Models** – OvalEdge now supports configuring chat models through Amazon Bedrock for various AI-powered capabilities, including:
* askEdgi analysis.
* Recipe creation.
* Automatic generation of business and technical descriptions for objects.
* Term recommendations.
* Data Classification Recommendations.
Administrators can configure Amazon Bedrock using **AWS IAM Role ARN-based authentication** and a secure cross-account AWS configuration. OvalEdge can use **Anthropic Claude models through Amazon Bedrock** for supported AI capabilities.
The following Claude models are supported through Amazon Bedrock:
* **Claude Sonnet 4.5**
* **Claude Haiku 4.5**
* **Claude Opus 4.5**
Existing integrations with **OpenAI** and **Google Gemini** remain supported.
## Connectors
### New & Improved
**Improved QuickBase Profiling Without Legacy XML API Limitations**
The QuickBase connector used a legacy XML API for profiling, which required multiple API calls to retrieve metadata and data. This increased profiling time, added complexity, and made the implementation difficult to maintain and extend.
With this enhancement, QuickBase profiling no longer depends on the legacy XML API and uses a more efficient approach to retrieve the required profiling information. This reduces unnecessary API calls and improves profiling performance.
**Support User Token Authentication for Quickbase Connector**
The Quickbase connector supported only Username and Password authentication, limiting integrations to the legacy method and not supporting data profiling.
With this enhancement, the connector now supports User Token authentication as an alternative to Username and Password. Users can select the required authentication method and provide a Quickbase User Token, while existing Username and Password connections remain functional.
## Known Issues
### askEdgi
**Issue**:
Simple prompts submitted for a Pinned Table may take significantly longer to process and consume more input and output tokens than expected. The system may retrieve additional Business Glossary and Tags information instead of focusing primarily on the pinned object, increasing token consumption and response time. This behavior occurs with both Amazon Bedrock Claude models and OpenAI models.
**Impact**:
Users may experience response times of around 104 seconds for simple prompts and higher-than-expected token consumption when querying Pinned Tables.
**Workaround**:
No workaround is currently available. Until the issue is addressed in Release8.3, users may experience increased response times and token consumption when submitting prompts on Pinned Tables.
***
Copyright © 2026, OvalEdge LLC, Peachtree Corners, GA, USA.
---
# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.
## Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.
Perform an HTTP GET request on the current page URL with the `ask` query parameter, and the optional `goal` query parameter:
```
GET https://docs.ovaledge.com/release8.2/patch-releases/release8.2.1.md?ask=&goal=
```
`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is optional and describes the broader end goal you are ultimately trying to accomplish on behalf of the user. GitBook uses it to tailor the answer towards what is most useful for that goal.
The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.
Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.