Higress Engine Update

Overview of This Release

This release includes 42 updates, covering enhancements, bug fixes, performance optimizations, and more.

Distribution of Updates

• New Features: 21 items

• Bug Fixes: 14 items

• Refactoring Optimizations: 4 items

• Documentation Updates: 2 items

• Testing Improvements: 1 item

Highlights

This release includes 3 important updates that deserve attention:

• feat: add MCP SSE stateful session load balancer support (#2818): This feature enables MCP services based on the SSE protocol to better maintain persistent connections between clients and servers, enhancing user experience and application performance, especially in scenarios requiring long-lived connections for data push.

• feat: Support adding a proxy server in between when forwarding requests to upstream (#2710): This feature allows users to use a proxy server when forwarding requests to upstream services, enhancing system flexibility and security, suitable for scenarios requiring communication through specific proxies.

• feat(ai-proxy): add auto protocol compatibility for OpenAI and Claude APIs (#2810): By automatically detecting and converting protocols, all AI Providers can be compatible with both OpenAI and Claude protocols, enabling smooth integration with Claude Code.

For detailed information, please refer to the section below for important feature descriptions.

Important Feature Details

Below are detailed explanations of the important features and improvements in this release:

1. feat: add MCP SSE stateful session load balancer support

Related PR: #2818 | Contributor: @johnlanni

Background

As the demand for real-time communication grows, Server-Sent Events (SSE) has become a critical technology for many applications. However, in distributed systems, ensuring that requests from the same client are always routed to the same backend service to maintain session state has become a challenge. Traditional load balancing strategies cannot meet this demand. This feature addresses this issue by introducing MCP SSE stateful session load balancing support. By specifying the mcp-sse type in the higress.io/load-balance annotation, users can easily implement state session management for SSE connections. The target user group mainly consists of application developers and service providers who need to perform real-time data pushing in a distributed environment.

Feature Details

This PR primarily implements the following functionalities:

1. Extended load-balance annotation: Support for the mcp-sse value has been added in the loadbalance.go file, and the McpSseStateful field has been added to the LoadBalanceConfig struct.

2. Simplified Configuration: Users only need to set mcp-sse in the higress.io/load-balance annotation to enable the feature without extra configuration.

3. Backend Address Encoding: When MCP SSE stateful session load balancing is enabled, the backend address will be Base64 encoded and embedded into the session ID of the SSE message. This ensures that the client can correctly identify and maintain the session.
   The core technological innovation lies in dynamically generating SSE session-related configurations via EnvoyFilter, thereby realizing state session management.

How to Use

To use this feature, users need to follow these steps:

1. Enable the Feature: Add higress.io/load-balance: mcp-sse annotation in the Ingress resource.

2. Configuration Example:

apiVersion: networking.k8s.io/v1
kind: Ingress
metadata:
  name: sse-ingress
  annotations:
    higress.io/load-balance: mcp-sse
spec:
  rules:
  - host: example.com
    http:
      paths:
      - path: /mcp-servers/test/sse
        pathType: Prefix
        backend:
          service:
            name: sse-service
            port:
              number: 80

3. Testing: Access the SSE endpoint using the curl command and check if the returned message contains the correct session ID.
   Note:

• Ensure the backend service can handle Base64 encoded session IDs.

• Avoid frequent changes to the backend service deployment to prevent session consistency issues.

Feature Value

This feature provides users with the following specific benefits:

1. Session Consistency: Ensures that requests from the same client are always routed to the same backend service, thus maintaining consistency of session state.

2. Simplified Configuration: Enables the feature through simple annotation configuration, reducing configuration complexity for users.

3. Enhanced User Experience: Applications relying on SSE, such as real-time notifications and stock quotes, can offer a more stable and consistent service experience.

4. Reduced Operational Costs: Minimizes errors and failures caused by session inconsistencies, reducing the workload of the operations team.

2. feat: Support adding a proxy server in between when forwarding requests to upstream

Related PR: #2710 | Contributor: @CH3CHO

Background

In modern microservice architectures, especially in complex network environments, directly forwarding requests from clients to backend services may encounter various issues such as network security and performance bottlenecks. Introducing an intermediary proxy server can effectively address these issues, such as through the proxy server for traffic control, load balancing, SSL offloading, etc. Additionally, in some cases, enterprises may need to use specific proxy servers to meet compliance and security requirements. The target user group for this feature mainly consists of enterprises and developers looking to optimize request forwarding paths in complex network environments.

Feature Details

This PR primarily implements the configuration of one or more proxy servers within the McpBridge resource and allows specifying a designated proxy server for each registry. Specific implementations include: 1. Adding a proxies field in the McpBridge resource definition for configuring the proxy server list, as well as adding a proxyName field in the registries item to associate the proxy server with the registry. 2. When the McpBridge resource is created or updated, the system automatically generates corresponding EnvoyFilter resources based on the configuration that define how to forward requests to the specified proxy server. 3. Additionally, EnvoyFilter resources are generated for each service bound to a proxy, ensuring they can correctly point to the local listeners on the corresponding proxy server. The entire technical implementation is based on Envoy's advanced routing capabilities, showcasing the project's powerful functionality in dealing with complex network topologies.

How to Use

Enabling this feature requires first configuring at least one proxy server within the McpBridge resource. This can be done by adding a new ProxyConfig object to the spec.proxies array, where each object must include necessary information such as name, serverAddress, serverPort, etc. Then, for registry entries wishing to use the proxy server, simply reference the defined proxy name in their proxyName field. Once configured, the system will automatically handle all related EnvoyFilter generation tasks. It is important to carefully check the correctness of the configuration file before actual deployment to avoid issues like service unavailability due to erroneous configuration.

Feature Value

The newly added proxy server support greatly enhances the network flexibility of the system, allowing users to flexibly adjust the request forwarding paths according to their needs. For instance, by setting different proxy servers, data transmission optimization between multiple regions can be easily achieved; additionally, with the extra security features provided by the proxy layer (such as SSL encryption), the overall security of the system is significantly improved. Furthermore, this functionality also helps simplify operational management, especially in cases where frequent adjustments to the network architecture are needed, allowing for quick responses to changes through simple configuration changes without making major modifications to the underlying infrastructure. In summary, this improvement not only expands the applicability of the project but also provides users with more powerful tools to tackle increasingly complex network challenges.

3. feat(ai-proxy): add auto protocol compatibility for OpenAI and Claude APIs

Related PR: #2810 | Contributor: @johnlanni

Background

In the AI proxy plugin, users may need to interact with multiple AI service providers (such as OpenAI and Anthropic Claude) simultaneously. These providers often use different API protocols, leading to complexity and an increased chance of error when users switch services as they need to manually configure protocol types. This feature addresses this issue by allowing users to seamlessly utilize services from different providers without worrying about the underlying protocol differences. The target user group is developers and enterprises looking to simplify the AI service integration process.

Feature Details

This PR implements automatic protocol compatibility, with the core technological innovation lying in automatically detecting request paths and intelligently conducting protocol conversions based on the capabilities of the target provider. Specifically, when the request path is /v1/chat/completions, it is recognized as the OpenAI protocol; when the request path is /v1/messages, it is recognized as the Claude protocol. If the target provider does not support the native Claude protocol, the plugin will convert requests from Claude format to OpenAI format, and vice versa. In the main.go file, logic for automatic protocol detection based on request paths has been added, and path replacements will be made if necessary. Additionally, a claude_to_openai.go file has been added to implement the specific conversion logic from Claude to OpenAI protocol.

How to Use

Enabling this feature is straightforward; users simply send requests as usual without extra configuration. For example, for requests in the OpenAI protocol, the URL is http://your-domain/v1/chat/completions, whereas for requests in the Claude protocol, the URL is http://your-domain/v1/messages. The plugin will automatically detect and handle protocol conversions. If the target provider does not support Claude protocol, the plugin will convert it into OpenAI format. Example configurations are as follows:

provider:
  type: claude  # 原生支持Claude协议的供应商
  apiTokens:
    - 'YOUR_CLAUDE_API_TOKEN'
  version: '2023-06-01'

Note: Ensure API tokens and version numbers are correctly configured so that the plugin can properly identify and handle requests.

Feature Value

This feature significantly enhances the ease of use and flexibility of the AI proxy plugin, reducing the user's configuration burden. Through automatic protocol detection and intelligent conversion, users can more easily switch between different AI service providers without worrying about protocol compatibility issues. This not only improves development efficiency but also enhances the stability and reliability of the system. Moreover, this feature supports streaming responses, further expanding its application scenarios, especially in real-time interactive contexts. In conclusion, this feature provides users with a more efficient and simpler way to integrate and manage multiple AI service providers.

Complete Change Log

New Features (Features)

• Related PR: #2847
  Contributor: @Erica177
  Change Log: This PR adds a secure mode to Nacos MCP, involving modifications to the mcp_model.go and watcher.go files, including new and adjusted configuration items.
  Feature Value: By adding secure mode support, the security of Nacos MCP services is enhanced, allowing users to manage their microservice configurations in a more secure environment.

• Related PR: #2842
  Contributor: @hanxiantao
  Change Log: Added detailed documentation in Chinese and English for the hmac-auth-apisix plugin and corresponding test cases to ensure the stability and reliability of the newly added features.
  Feature Value: By increasing documentation and testing, the usability and stability of the hmac-auth-apisix plugin have been improved, helping users better understand and use the HMAC authentication mechanism, thereby enhancing API security.

• Related PR: #2823
  Contributor: @johnlanni
  Change Log: Added OpenRouter as an AI service provider, supporting unified API access to multiple AI models. Core implementations include support for chat completions and text completions.
  Feature Value: By introducing OpenRouter, users can flexibly choose different AI models and interact with them, simplifying the complexity of using AI services across platforms and enhancing the user experience.

• Related PR: #2815
  Contributor: @hanxiantao
  Change Log: This PR adds the hmac-auth-apisix plugin, implementing authentication for API requests. It uses the HMAC algorithm to generate signatures for validating the integrity and authenticity of requests.
  Feature Value: The newly added hmac-auth-apisix plugin enhances system security by ensuring that only authenticated clients can access protected resources, improving user experience and system protection capabilities.

• Related PR: #2808
  Contributor: @daixijun
  Change Log: Added support for Anthropic API and OpenAI v1/models interface, expanding the compatibility and functionality of DeepSeek.
  Feature Value: The new support enables users to utilize more AI service options, enhancing the system's flexibility and practicality.

• Related PR: #2805
  Contributor: @johnlanni
  Change Log: Added a JSON-RPC protocol conversion plugin, capable of extracting request and response information to headers from the MCP protocol, facilitating further observation, rate limiting, authentication, etc.
  Feature Value: This feature allows users to use JSON-RPC for higher-level policy control in A2A protocols, such as authentication and traffic management, thereby enhancing the system's flexibility and security.

• Related PR: #2788
  Contributor: @zat366
  Change Log: This PR updates the dependencies in mcp-server to support responding to images with MCP plugins. Achieved through updates to go.mod and go.sum files.
  Feature Value: The new feature allows MCP plugins to handle and respond to image data, enhancing the system's multimedia processing capabilities and providing users with richer content display options.

• Related PR: #2769
  Contributor: @github-actions[bot]
  Change Log: This PR updates CRD files in the helm folder to add new property definitions regarding proxies.
  Feature Value: By adding new properties through the updated CRD files, the richness and completeness of Kubernetes resource definitions are enhanced, improving the flexibility and scalability of system configurations.

• Related PR: #2761
  Contributor: @johnlanni
  Change Log: This PR introduces two new deduplication strategies: SPLIT_AND_RETAIN_FIRST and SPLIT_AND_RETAIN_LAST, respectively used to retain the first and last elements of comma-separated header values.
  Feature Value: The new strategies provide users with finer control options, allowing them to choose to retain specific positions of data during deduplication, thus better meeting diverse needs.

• Related PR: #2739
  Contributor: @WeixinX
  Change Log: Added a plugin configuration field reroute, allowing users to control whether to disable route reselection. This feature was implemented by modifying the main configuration file and adding related test cases.
  Feature Value: This feature provides a way for users to finely control routing behavior during request processing, enhancing system flexibility and configurability to meet specific scene requirements.

• Related PR: #2730
  Contributor: @rinfx
  Change Log: This PR adds tool usage support for Bedrock services by modifying structures and logic in bedrock.go and other files, enabling the system to handle tool-related requests.
  Feature Value: The newly added functionality allows users to effectively utilize tool calling capabilities in the Bedrock environment, enhancing system flexibility and functionality, better meeting application scenario requirements that integrate external tools.

• Related PR: #2729
  Contributor: @rinfx
  Change Log: This PR adds a length limit for each value in the AI statistics plugin, automatically truncating when the set length is exceeded. This helps reduce memory usage when processing large files (such as base64 encoded images, videos).
  Feature Value: By limiting and truncating overly long data values, this feature can effectively prevent memory overflow problems caused by handling large media files, thereby improving system stability and performance.

• Related PR: #2713
  Contributor: @Aias00
  Change Log: This PR adds Grok provider support to the AI proxy, including new Grok Go file implementations and updates to related documentation.
  Feature Value: By integrating Grok as a new AI provider, users can now leverage Grok's AI capabilities to handle requests, increasing system flexibility and functional diversity.

• Related PR: #2712
  Contributor: @SCMRCORE
  Change Log: Added support for Gemini model thinking functionality, particularly adapting for the 2.5 Flash, 2.5 Pro, and 2.5 Flash-Lite models.
  Feature Value: The enhanced functionality of the AI proxy plugin allows users to utilize specific Gemini models for more complex thinking tasks, improving user experience and application range.

• Related PR: #2704
  Contributor: @hanxiantao
  Change Log: This PR implements Rust WASM plugin support for Redis database configuration options and improves demo-wasm to obtain Redis configuration from the Wasm plugin configuration.
  Feature Value: This functionality allows developers greater flexibility in configuring and integrating Redis databases when using Rust WASM plugins, improving development efficiency and application configurability.

• Related PR: #2698
  Contributor: @erasernoob
  Change Log: Achieves multi-modal support for Gemini models, adding the ability to process images and text. Enhanced functionality through the introduction of new dependencies and modifications to existing code logic.
  Feature Value: Enhances the functionality of the AI proxy plugin, enabling it to support more complex multi-modal data handling, providing users with richer and more flexible AI service experiences.

• Related PR: #2696
  Contributor: @rinfx
  Change Log: This PR added streaming response support when enabling the content safety plugin, adjusting the detection frequency through the bufferLimit parameter to improve the flexibility and efficiency of content detection.
  Feature Value: The added streaming response functionality allows users to handle content safety detection more efficiently, reducing latency and enhancing user experience, especially suitable for applications requiring real-time feedback.

• Related PR: #2671
  Contributor: @Aias00
  Change Log: Implements path suffix and content type filtering functionality to address performance and resource management issues in the ai-statistics plugin. Introduced the SkipProcessing mechanism to avoid indiscriminate processing of all requests, reducing unnecessary response body caching.
  Feature Value: Enhances the selective processing capability of the AI statistics plugin, improving system performance and optimizing resource usage efficiency, particularly beneficial for large and complex API request scenarios, which can significantly improve user experience.

Bug Fixes

• Related PR: #2816
  Contributor: @Asnowww
  Change Log: This PR fixes a spelling error in the scanners-user-agents.data file, correcting 'scannr' to 'scanner'.
  Feature Value: Correcting spelling errors in documents can enhance document accuracy and readability, aiding users to better understand and utilize related features.

• Related PR: #2799
  Contributor: @erasernoob
  Change Log: Fixed the build command of the wasm-go-build plugin to ensure that all files in the directory are included during compilation, resolving compilation failure issues caused by dependencies.
  Feature Value: By fixing the build command, errors due to missing necessary files during compilation are avoided, enhancing the stability and reliability of the build process, providing a better development experience for developers.

• Related PR: #2787
  Contributor: @co63oc
  Change Log: Fixed a spelling error in the RegisteTickFunc function, ensuring the correctness of the timer task registration function. By correcting the key function name, potential functional failures are avoided.
  Feature Value: Fixing the error caused by a typo in timer tasks ensures that they work normally, enhancing system stability and reliability, thereby ensuring applications dependent on timer task execution run as expected.

• Related PR: #2786
  Contributor: @CH3CHO
  Change Log: This PR removed the 'accept-encoding' header when processing SSE transmission requests with the mcp-session filter, resolving data parsing errors caused by attempting to handle compressed response body data correctly.
  Feature Value: The fix ensures that the MCP server operates normally when using SSE transmissions upstream, avoiding data parsing errors caused by compression, improving the system's stability and reliability.

• Related PR: #2782
  Contributor: @CH3CHO
  Change Log: This PR fixes issues where the Azure URL configuration component was inadvertently changed by defining a new enumeration type azureServiceUrlType to ensure URL component correctness and consistency.
  Feature Value: The fix ensures that users can maintain their original Azure service URL configurations when using the AI proxy, avoiding service call failures or inconsistencies caused by incorrect changes.

• Related PR: #2757
  Contributor: @Jing-ze
  Change Log: Fixed issues in building Envoy filter unit tests in mcp server, ensuring correctness and stability of the test cases.
  Feature Value: By fixing errors in unit tests, the reliability and maintainability of the code is enhanced, helping developers better perform subsequent development and debugging work.

• Related PR: #2755
  Contributor: @CH3CHO
  Change Log: This PR fixes an issue that threw an error when adding duplicate IPs in the ip-restriction configuration by ignoring errors for existing IPs and displaying specific error details returned by iptree.
  Feature Value: Allowing duplicates in the IP restriction list improves configuration flexibility and user experience while ensuring that other types of errors are still handled effectively.

• Related PR: #2754
  Contributor: @Jing-ze
  Change Log: Corrected stopping and buffering issues when decoding data in golang-filter, ensuring a more stable data processing flow.
  Feature Value: The fix resolves errors in the data decoding process, enhancing system reliability and user experience by preventing potential data loss or processing anomalies.

• Related PR: #2743
  Contributor: @Jing-ze
  Change Log: Fixed an error that occurred when setting ip_source_type to origin-source, ensuring the IP restriction function could be configured correctly based on the source type.
  Feature Value: This fix resolves incorrect IP source type settings under specific conditions, enhancing system stability and security, allowing users to reliably utilize IP restriction features.

• Related PR: #2723
  Contributor: @CH3CHO
  Change Log: Fixed the _match_service_ rule issue in the C++ Wasm plugin due to using incorrect attribute names, restoring the rule to normal by modifying to the correct attribute name.
  Feature Value: This fix resolves service routing issues due to matching rule errors, enhancing system stability and accuracy, ensuring users can correctly access the desired services.

• Related PR: #2706
  Contributor: @WeixinX
  Change Log: Fixed an issue where the transformer would perform addition operations when replacing keys did not exist and added test cases for mapping operations to ensure correct conversions from headers/querys to body and vice versa.
  Feature Value: This fix enhances system stability and reliability, preventing erroneous data operations, giving users greater confidence in data processing logic, and improving user experience.

• Related PR: #2663
  Contributor: @CH3CHO
  Change Log: Fixed errors in the name escaping logic for the bedrock model, removed unnecessary URL encoding handling from the request body, and ensured responses align with expectations.
  Feature Value: Fixing the name escaping logic issues improves system stability and compatibility, ensuring that users do not encounter problems due to mismatched escapes during usage.

• Related PR: #2653
  Contributor: @johnlanni
  Change Log: Fixed an issue where the AI routing fallback functionality failed when using Bedrock. Ensured that paths could be correctly obtained even when headers are nil to avoid nil pointer exceptions.
  Feature Value: This fix resolves request rejection issues due to signature verification failures under specific conditions, enhancing system stability and reliability, ensuring users can smoothly access services.

• Related PR: #2628
  Contributor: @co63oc
  Change Log: This PR corrected spelling errors in multiple files, involving modifications across 5 files and 36 lines of code, ensuring the accuracy of documentation and comments.
  Feature Value: Correcting spelling errors improves the professionalism of the codebase, enabling developers to accurately understand content while reading documentation, thereby reducing errors caused by misunderstanding.

Refactoring

• Related PR: #2777
  Contributor: @StarryVae
  Change Log: Updated the ai-prompt-decorator plugin to the new encapsulated API, improving the initialization configuration and request header handling method calls.
  Feature Value: This refactor improves code consistency and maintainability, making it easier for developers to integrate and use ai-prompt-decorator functionalities.

• Related PR: #2773
  Contributor: @CH3CHO
  Change Log: Refactored the path to API name mapping logic in ai-proxy, introducing regular expressions to simplify the mapping process and added test cases to validate functionality correctness.
  Feature Value: By optimizing the structure of the path mapping logic, improved code maintainability and scalability, making it easier to support more paths, indirectly enhancing system flexibility and user experience.

• Related PR: #2740
  Contributor: @CH3CHO
  Change Log: This PR lowers some log levels in the ai-statistics component from warn to info to more accurately reflect the actual importance of these log messages.
  Feature Value: By adjusting log levels, log records better align with actual needs, helping to reduce false alarm rates for users when checking logs, improving user experience.

• Related PR: #2711
  Contributor: @johnlanni
  Change Log: This PR deprecated the approach of using slashes as connectors in the mcp server and tool, changing to a format that complies better with function naming conventions. This includes updating dependency versions in some codebase and adjusting related files.
  Feature Value: By adhering to standard function naming conventions, this change enhances code consistency and readability, helping to reduce future maintenance costs and minimizing potential errors due to non-compliant naming.

Documentation

• Related PR: #2770
  Contributor: @co63oc
  Change Log: Corrected spelling errors in multiple files, including test files, README, and variable names and configuration item names in Go code.
  Feature Value: Improved the accuracy and readability of the documentation, ensuring consistency in code and user experience. These changes help avoid confusion or configuration issues arising from spelling errors for users using this plugin.

Testing

• Related PR: #2809
  Contributor: @Jing-ze
  Change Log: Added unit tests for multiple Wasm extensions and introduced a CI/CD workflow to automate these tests, ensuring code quality and stability.
  Feature Value: Improved reliability of the Wasm plugin through comprehensive unit testing and automated CI/CD processes, helping developers discover and fix issues faster, enhancing user experience.

Release Statistics

• New Features: 21 items

• Bug Fixes: 14 items

• Refactoring Optimizations: 4 items

• Documentation Updates: 2 items

• Testing Improvements: 1 item

Total: 42 changes (including 3 important updates)

Higress Console

Overview of This Release

This release includes 12 updates, covering enhancements, bug fixes, performance optimizations, and more.

Distribution of Updates

• New Features: 5 items

• Bug Fixes: 5 items

• Refactoring Optimizations: 2 items

Complete Change Log

New Features (Features)

• Related PR: #585
  Contributor: @johnlanni
  Change Log: This PR adds new AI service providers and updates the list of available models, including updates to translation files to support the newly added providers.
  Feature Value: By introducing more AI service providers and updating the model list, users can now access a wider range of service options, enhancing system flexibility and practicality.

• Related PR: #582
  Contributor: @Thomas-Eliot
  Change Log: Added support for ai-load-balancer plugin, enabling visual configuration in higress-console and defining its priority within the system.
  Feature Value: By providing a blank screen configuration option, user management efficiency and flexibility for the AI load balancer is greatly enhanced, lowering the usage threshold.

• Related PR: #579
  Contributor: @JayLi52
  Change Log: This update adds support for PostgreSQL and ClickHouse databases in the MCP server management features, while optimizing MySQL database connection string format and fixing various database connection related issues.
  Feature Value: The newly added database support expands the application range of the MCP, allowing users to flexibly choose database types that fit their needs, improving system compatibility and user experience.

• Related PR: #572
  Contributor: @CH3CHO
  Change Log: This PR adds management features for proxy servers, including new classes and service controllers, allowing users to configure and manage proxy servers.
  Feature Value: With the new support added, users can manage and configure proxy servers more flexibly, improving system flexibility and availability.

• Related PR: #565
  Contributor: @Aias00
  Change Log: This PR improves MCP server management tasks 6 and 7, including updates to README.md documentation, modifying system service implementation code, and optimizing ConfigMap processing logic.
  Feature Value: By improving MCP server management functionality, the system's stability and maintainability are enhanced, simplifying user management of Higress configurations and improving user experience.

Bug Fixes

• Related PR: #584
  Contributor: @CH3CHO
  Change Log: Fixed errors that occurred when authentication was enabled but no allowed consumers were present, including incorrectly clearing the allowed consumers list and displaying incorrect authentication status.
  Feature Value: Ensures that the authentication function works correctly even without allowed consumers and that the user interface accurately reflects the current authentication status.

• Related PR: #581
  Contributor: @hongzhouzi
  Change Log: Fixed NPE exceptions that occurred when updating the OpenAPI MCP server and corrected PostgreSQL enumeration values to ensure consistency with constants in Higress.
  Feature Value: Improved system stability and reliability by resolving NPE issues, while the consistency of enumeration values enhances the accuracy of configuration management, reducing potential sources of error.

• Related PR: #577
  Contributor: @CH3CHO
  Change Log: Synchronized the front-end and back-end domain name validation regular expressions to ensure long top-level domains like test.internal are accepted, involving minor code modifications and added test cases.
  Feature Value: Fixed issues where valid domain names were not accepted due to inconsistencies in validation rules used by the front-end and back-end, enhancing system compatibility and user experience.

• Related PR: #574
  Contributor: @CH3CHO
  Change Log: Fixed logical errors while filtering V1alpha1WasmPlugin by internal flag to ensure that non-internal instances are not mistakenly returned.
  Feature Value: Enhanced system accuracy ensures that users retrieve the correct list of plugin instances, avoiding data inconsistencies caused by logical errors.

• Related PR: #570
  Contributor: @CH3CHO
  Change Log: Corrected a spelling error that caused a 'Cannot read properties of undefined' error when editing OpenAI type LLM providers.
  Feature Value: By fixing this issue, the risk of runtime errors when configuring OpenAI service providers is avoided, improving system stability and user experience.

Refactoring

• Related PR: #573
  Contributor: @CH3CHO
  Change Log: Refactored the authentication module integrated into the MCP server, allowing common routing and MCP servers to share the same authentication logic. Major changes include adding, deleting, and modifying code across multiple files.
  Feature Value: Through refactoring the authentication module, unified authentication logic is achieved, enhancing code maintainability and reusability, reducing duplicate code, and helping to improve overall system stability and performance.

• Related PR: #571
  Contributor: @JayLi52
  Change Log: By updating the import method for the Monaco editor and configuring on-demand loading, optimized performance in EditToolDrawer, McpServerCommand, and MCPDetail components.
  Feature Value: Improved application loading speed and response efficiency, reducing unnecessary resource consumption and enhancing user experience.

Release Statistics

• New Features: 5 items

• Bug Fixes: 5 items

• Refactoring Optimizations: 2 items

Total: 12 changes

Thank you to all contributors for your hard work!