feat: Support the MCP client and dynamically fetch the model list by lhpqaq · Pull Request #298 · apache/bigtop-manager

lhpqaq · 2026-03-26T14:21:48Z

Support adding MCP servers of the local, SSE, and streamHttp types
Support dynamically fetching the model list
Support displaying the tool invocation process

- Deleted ClusterFunctions, HostFunctions, StackFunctions, AIServiceToolsProvider, and InfoToolsProvider classes as they are no longer needed. - Removed associated test classes: ClusterFunctionsTest, HostFunctionsTest, StackFunctionsTest, and AIServiceToolsProviderTest. - Updated application.yml to include new client configuration. - Added new API methods in llm-config for fetching platform models. - Enhanced UI components to support model fetching with loading states and error handling. - Updated localization files for new UI strings related to model fetching. - Added repository for Spring milestones in pom.xml.

Copilot

Pull request overview

This PR expands the AI assistant feature set by adding MCP client support (local/SSE/streamable HTTP), enabling dynamic model discovery from providers, and surfacing tool invocation progress in the UI during streaming responses.

Changes:

Add MCP client configuration/initialization and wire MCP tool callbacks into Spring AI assistants with tool-execution event streaming.
Add a backend API to fetch models dynamically and update UI to refresh model options on demand.
Update AI assistant UI/store to display tool execution steps alongside the streaming assistant message.

Reviewed changes

Copilot reviewed 42 out of 42 changed files in this pull request and generated 11 comments.

Show a summary per file

File	Description
pom.xml	Adds Spring milestone repository configuration.
bigtop-manager-ui/src/store/llm-config/index.ts	Adds store action to refresh platform models.
bigtop-manager-ui/src/store/ai-assistant/index.ts	Tracks tool executions + pending streaming AI record.
bigtop-manager-ui/src/pages/system-manage/llm-config/components/add-llm-item.vue	Adds “Fetch Models” button to model selector.
bigtop-manager-ui/src/locales/zh_CN/llm-config.ts	Adds new i18n strings for model fetching.
bigtop-manager-ui/src/locales/en_US/llm-config.ts	Adds new i18n strings for model fetching.
bigtop-manager-ui/src/features/ai-assistant/index.vue	Renders streaming tool/AI records during SSE.
bigtop-manager-ui/src/features/ai-assistant/chat-message.vue	Adds UI rendering for tool execution records.
bigtop-manager-ui/src/api/llm-config/types.ts	Adds request type for platform model fetch.
bigtop-manager-ui/src/api/llm-config/index.ts	Adds API call for platform model list endpoint.
bigtop-manager-ui/src/api/ai-assistant/types.ts	Extends message/event typing for tool execution fields.
bigtop-manager-server/src/test/java/org/apache/bigtop/manager/server/tools/provider/AIServiceToolsProviderTest.java	Removes tests for removed langchain4j tool provider.
bigtop-manager-server/src/test/java/org/apache/bigtop/manager/server/tools/functions/StackFunctionsTest.java	Removes tests for removed langchain4j stack functions.
bigtop-manager-server/src/test/java/org/apache/bigtop/manager/server/tools/functions/HostFunctionsTest.java	Removes tests for removed langchain4j host functions.
bigtop-manager-server/src/test/java/org/apache/bigtop/manager/server/tools/functions/ClusterFunctionsTest.java	Removes tests for removed langchain4j cluster functions.
bigtop-manager-server/src/main/resources/application.yml	Adds MCP client configuration + defaults.
bigtop-manager-server/src/main/java/org/apache/bigtop/manager/server/tools/provider/InfoToolsProvider.java	Removes langchain4j ToolProvider implementation.
bigtop-manager-server/src/main/java/org/apache/bigtop/manager/server/tools/provider/AIServiceToolsProvider.java	Removes tool provider selector used previously by chatbot.
bigtop-manager-server/src/main/java/org/apache/bigtop/manager/server/tools/functions/StackFunctions.java	Removes langchain4j tool functions.
bigtop-manager-server/src/main/java/org/apache/bigtop/manager/server/tools/functions/HostFunctions.java	Removes langchain4j tool functions.
bigtop-manager-server/src/main/java/org/apache/bigtop/manager/server/tools/functions/ClusterFunctions.java	Removes langchain4j tool functions.
bigtop-manager-server/src/main/java/org/apache/bigtop/manager/server/service/impl/LLMConfigServiceImpl.java	Adds dynamic model fetch + uses it in model validation.
bigtop-manager-server/src/main/java/org/apache/bigtop/manager/server/service/impl/ChatbotServiceImpl.java	Streams tool execution events via SSE and improves emitter lifecycle handling.
bigtop-manager-server/src/main/java/org/apache/bigtop/manager/server/service/LLMConfigService.java	Adds platformModels service API.
bigtop-manager-server/src/main/java/org/apache/bigtop/manager/server/model/vo/TalkVO.java	Extends SSE payload with tool execution fields.
bigtop-manager-server/src/main/java/org/apache/bigtop/manager/server/model/converter/AuthPlatformConverter.java	Filters null credential keys before toMap conversion.
bigtop-manager-server/src/main/java/org/apache/bigtop/manager/server/mcp/tool/StackMcpTool.java	Adds logging when MCP tool is invoked.
bigtop-manager-server/src/main/java/org/apache/bigtop/manager/server/controller/LLMConfigController.java	Adds `/platforms/{platformId}/models` endpoint.
bigtop-manager-server/pom.xml	Removes direct langchain4j dependency from server module.
bigtop-manager-bom/pom.xml	Bumps Spring AI version; adds MCP client starter to BOM.
bigtop-manager-ai/src/test/java/org/apache/bigtop/manager/ai/config/McpAsyncClientManagerTest.java	Adds tests for MCP connection parsing behavior.
bigtop-manager-ai/src/test/java/assistant/GeneralAssistantFactoryTest.java	Updates tests for revised factory method signatures.
bigtop-manager-ai/src/main/java/org/apache/bigtop/manager/ai/platform/QianFanAssistant.java	Adds dynamic model fetch + MCP tool callback observation.
bigtop-manager-ai/src/main/java/org/apache/bigtop/manager/ai/platform/OpenAIAssistant.java	Adds MCP tool callback observation + base URL/model streaming tweaks.
bigtop-manager-ai/src/main/java/org/apache/bigtop/manager/ai/platform/DeepSeekAssistant.java	Adds MCP tool callback observation + streaming safety tweaks.
bigtop-manager-ai/src/main/java/org/apache/bigtop/manager/ai/platform/DashScopeAssistant.java	Adds MCP tool callback observation + streaming safety tweaks.
bigtop-manager-ai/src/main/java/org/apache/bigtop/manager/ai/core/factory/AIAssistantFactory.java	Adjusts factory API; adds model discovery and tool-event hook.
bigtop-manager-ai/src/main/java/org/apache/bigtop/manager/ai/core/factory/AIAssistant.java	Adds MCP client + tool-event listener support and event record type.
bigtop-manager-ai/src/main/java/org/apache/bigtop/manager/ai/core/AbstractAIAssistant.java	Implements model discovery helper and tool-execution event emission wiring.
bigtop-manager-ai/src/main/java/org/apache/bigtop/manager/ai/config/McpAsyncClientManager.java	Adds MCP async client manager with connection parsing and initialization.
bigtop-manager-ai/src/main/java/org/apache/bigtop/manager/ai/assistant/GeneralAssistantFactory.java	Wires MCP clients + tool execution listener; adds model discovery method.
bigtop-manager-ai/pom.xml	Adds MCP client starter dependency.

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

Copilot · 2026-03-29T08:59:09Z

pom.xml

+    <repositories>
+        <repository>
+            <snapshots>
+                <enabled>false</enabled>
+            </snapshots>
+            <id>spring-milestones</id>
+            <name>Spring Milestones</name>
+            <url>https://repo.spring.io/milestone</url>
+        </repository>
+    </repositories>


The added Spring milestone repository appears unnecessary for Spring AI 1.1.3 / spring-ai-starter-mcp-client-webflux (both are published to Maven Central). Keeping an extra repository can make builds less reproducible and increases supply-chain surface; consider removing it (or gating it behind a profile only when needed).

Copilot · 2026-03-29T08:59:09Z

...er-server/src/main/java/org/apache/bigtop/manager/server/controller/LLMConfigController.java

+    @Operation(summary = "list platform models", description = "List models from /v1/models")
+    @PostMapping("/platforms/{platformId}/models")
+    public ResponseEntity<List<String>> platformModels(
+            @PathVariable(name = "platformId") Long platformId, @RequestBody AuthPlatformReq authPlatformReq) {
+        AuthPlatformDTO authPlatformDTO = AuthPlatformConverter.INSTANCE.fromReq2DTO(authPlatformReq);
+        Map<String, String> authCredentials = authPlatformDTO.getAuthCredentials();
+        return ResponseEntity.success(llmConfigService.platformModels(platformId, authCredentials));
+    }


The new /platforms/{platformId}/models endpoint accepts AuthPlatformReq, but platformId is already provided in the path and the request type contains many unrelated fields. Consider introducing a dedicated request DTO (e.g., { authCredentials: [...] }) to keep the API contract clear and avoid implying that other fields are required/used.

Copilot · 2026-03-29T08:59:09Z

...r-server/src/main/java/org/apache/bigtop/manager/server/service/impl/ChatbotServiceImpl.java

+        AIAssistant aiAssistant = prepareTalk(threadId, command, emitter, emitterClosed);

        Flux<String> stringFlux =
                (command == null) ? aiAssistant.streamAsk(message) : Flux.just(aiAssistant.ask(message));


Flux.just(aiAssistant.ask(message)) evaluates ask() eagerly (before subscription). With tool-execution events now emitted during ask(), this can send SSE events before the emitter lifecycle handlers are registered / before the client is fully connected. Use a deferred publisher (e.g., Flux.defer/Mono.fromCallable) for the non-streaming path so execution starts on subscribe and can be cancelled/cleaned up consistently.

Suggested change

(command == null) ? aiAssistant.streamAsk(message) : Flux.just(aiAssistant.ask(message));

(command == null)

? aiAssistant.streamAsk(message)

: Flux.defer(() -> Flux.just(aiAssistant.ask(message)));

Copilot · 2026-03-29T08:59:10Z

bigtop-manager-ai/src/main/java/org/apache/bigtop/manager/ai/platform/OpenAIAssistant.java

+            OpenAiApi openAiApi = OpenAiApi.builder()
+                    .baseUrl(resolveDefaultBaseUrl())
+                    .apiKey(apiKey)
+                    .build();
+            OpenAiChatOptions.Builder optionsBuilder =
+                    OpenAiChatOptions.builder().model(model);
+            List<io.modelcontextprotocol.client.McpAsyncClient> mcpClients = getMcpAsyncClients();
+            if (!mcpClients.isEmpty()) {
+                optionsBuilder.toolCallbacks(buildObservedToolCallbacks(mcpClients));
+            }
+            OpenAiChatOptions options = optionsBuilder.build();


OpenAIAssistant supports a credential-provided baseUrl for model discovery, but chat requests still always use resolveDefaultBaseUrl() (ignoring that credential). This can make model listing succeed against a custom endpoint while chat requests go to a different host. Consider using the same base-url resolution for both chat and model listing to keep behavior consistent.

Copilot · 2026-03-29T08:59:10Z

...server/src/main/java/org/apache/bigtop/manager/server/service/impl/LLMConfigServiceImpl.java

+        if (supportModels == null || supportModels.isBlank()) {
+            return Collections.emptyList();
+        }
+        return List.of(supportModels.split(","));


getDefaultModels returns List.of(supportModels.split(",")) without trimming/filtering blanks. If supportModels contains spaces or trailing commas, callers may see invalid model IDs (e.g., " gpt-4" or ""). Consider trimming each entry and filtering empty strings before returning.

Suggested change

return List.of(supportModels.split(","));

String[] rawModels = supportModels.split(",");

List<String> models = new ArrayList<>();

for (String rawModel : rawModels) {

if (rawModel == null) {

continue;

}

String trimmed = rawModel.trim();

if (!trimmed.isEmpty()) {

models.add(trimmed);

}

}

if (models.isEmpty()) {

return Collections.emptyList();

}

return models;

Copilot · 2026-03-29T08:59:11Z

bigtop-manager-ui/src/store/ai-assistant/index.ts

+      const prevRecord = toolExecutions.value.find((execution) => execution.executionId === executionId)
+      const record = {
+        sender: 'AI' as const,
+        message: '',
+        messageType: 'tool' as const,
+        executionId,
+        toolName,
+        toolStatus,
+        toolPayload,
+        toolInput: toolStatus === 'started' ? toolPayload || '' : prevRecord?.toolInput || '',
+        toolOutput: toolStatus === 'completed' ? toolPayload || '' : prevRecord?.toolOutput || '',
+        toolError: toolStatus === 'failed' ? toolPayload || '' : prevRecord?.toolError || '',
+        createTime: dayjs().format()
+      }


updateToolExecution overwrites createTime on every tool status update. This can reorder tool records / show misleading timestamps (e.g., completed time instead of started time). Preserve the original createTime from prevRecord when present and only set it when creating a new execution record.

Copilot · 2026-03-29T08:59:11Z

bigtop-manager-ui/src/features/ai-assistant/chat-message.vue

+              <div class="tool-section">
+                <span class="tool-section-title">Input</span>
+                <markdown-view :mark-raw="toolInput" />
+              </div>
+              <div v-if="props.record.toolStatus === 'completed'" class="tool-section">
+                <span class="tool-section-title">Output</span>
+                <markdown-view :mark-raw="toolOutput" />
+              </div>
+              <div v-if="props.record.toolStatus === 'failed'" class="tool-section">
+                <span class="tool-section-title">Error</span>
+                <markdown-view :mark-raw="toolError" />


This component introduces new user-visible labels ("Input", "Output", "Error") and displays raw status strings (started/completed/failed) without i18n. Other ai-assistant components use useI18n(), so these should be translated and status values mapped to localized labels.

Copilot · 2026-03-29T08:59:11Z

...er/src/main/java/org/apache/bigtop/manager/server/model/converter/AuthPlatformConverter.java

+                .filter(item -> item != null && item.getKey() != null)
                .collect(Collectors.toMap(AuthCredentialReq::getKey, AuthCredentialReq::getValue));


Collectors.toMap(AuthCredentialReq::getKey, AuthCredentialReq::getValue) will throw a NullPointerException if any credential has a null value (e.g., frontend sends {key: "apiKey"} when value is undefined). Filter out null values (and ideally handle duplicate keys via a merge function) to avoid 500s on the new models endpoint and existing auth-platform endpoints.

Suggested change

.filter(item -> item != null && item.getKey() != null)

.collect(Collectors.toMap(AuthCredentialReq::getKey, AuthCredentialReq::getValue));

.filter(item -> item != null && item.getKey() != null && item.getValue() != null)

.collect(Collectors.toMap(AuthCredentialReq::getKey, AuthCredentialReq::getValue, (existing, replacement) -> replacement));

Copilot · 2026-03-29T08:59:11Z

bigtop-manager-server/src/main/resources/application.yml

+      client:
+        enabled: true
+        request-timeout-seconds: 120
+        init-timeout-seconds: 20
+        # Optional multi-connection definition.
+        # Format:
+        #   name=<id>,type=<sse|streamable-http|local>,baseUrl=<url>,endpoint=<path>,command=<cmd>,args=<a|b>,env=<K:V|K2:V2>,requestTimeoutSeconds=<n>,initTimeoutSeconds=<n>;
+        # Notes:
+        #   - sse / streamable-http use baseUrl (+ optional endpoint)
+        #   - local uses command/args/env
+        #   - requestTimeoutSeconds/initTimeoutSeconds are optional per-connection overrides
+        # Example:
+        # connections: >
+        #   name=embedded,type=sse,baseUrl=http://localhost:8080,endpoint=/mcp/sse;
+        #   name=remote,type=streamable-http,baseUrl=http://127.0.0.1:9000,endpoint=/mcp,requestTimeoutSeconds=180;
+        #   name=memory,type=local,command=npx,args=-y|@modelcontextprotocol/server-memory,env=NODE_ENV:production
+        connections: >
+            name=bm,type=sse,baseUrl=http://localhost:8080,endpoint=/mcp/sse


MCP client is enabled by default and ships with a default connection to http://localhost:8080. This can lead to unexpected outbound connections/timeouts in real deployments and makes the feature non-optional by default. Consider defaulting spring.ai.mcp.client.enabled to false and leaving connections empty (keep examples commented) so operators opt in explicitly.

Copilot · 2026-03-29T08:59:12Z

bigtop-manager-server/src/main/java/org/apache/bigtop/manager/server/mcp/tool/StackMcpTool.java

            stackVOList.add(stackVO);
        }
-
+        log.info("ListStacks tool called, total stacks: {}", stackVOList.size());


Logging every tool invocation at INFO (ListStacks tool called...) may be noisy if tools are invoked frequently during chats. Consider lowering this to DEBUG (or sampling) to avoid log volume growth in production.

Suggested change

log.info("ListStacks tool called, total stacks: {}", stackVOList.size());

log.debug("ListStacks tool called, total stacks: {}", stackVOList.size());

lhpqaq added 7 commits March 26, 2026 20:18

fix ci

e493960

fix ci

34277ae

feat: support multi-mcp

1418d5c

feat: Show toolExecutions in message

f5fef7e

mvn spotless:apply

1cdaa89

fix ci

02693ea

lhpqaq requested a review from Copilot March 29, 2026 08:50

Copilot started reviewing on behalf of lhpqaq March 29, 2026 08:51 View session

Copilot AI reviewed Mar 29, 2026

View reviewed changes

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

feat: Support the MCP client and dynamically fetch the model list#298

feat: Support the MCP client and dynamically fetch the model list#298
lhpqaq wants to merge 7 commits intoapache:mainfrom
lhpqaq:agent

lhpqaq commented Mar 26, 2026 •

edited

Loading

Uh oh!

Copilot AI left a comment

Uh oh!

Copilot AI Mar 29, 2026

Uh oh!

Copilot AI Mar 29, 2026

Uh oh!

Copilot AI Mar 29, 2026

Uh oh!

Copilot AI Mar 29, 2026

Uh oh!

Copilot AI Mar 29, 2026

Uh oh!

Copilot AI Mar 29, 2026

Uh oh!

Copilot AI Mar 29, 2026

Uh oh!

Copilot AI Mar 29, 2026

Uh oh!

Copilot AI Mar 29, 2026

Uh oh!

Copilot AI Mar 29, 2026

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants

-        return List.of(supportModels.split(","));
+        String[] rawModels = supportModels.split(",");
+        List<String> models = new ArrayList<>();
+        for (String rawModel : rawModels) {
+            if (rawModel == null) {
+                continue;
+            }
+            String trimmed = rawModel.trim();
+            if (!trimmed.isEmpty()) {
+                models.add(trimmed);
+            }
+        }
+        if (models.isEmpty()) {
+            return Collections.emptyList();
+        }
+        return models;

		.filter(item -> item != null && item.getKey() != null)
		.collect(Collectors.toMap(AuthCredentialReq::getKey, AuthCredentialReq::getValue));

	log.info("ListStacks tool called, total stacks: {}", stackVOList.size());
	log.debug("ListStacks tool called, total stacks: {}", stackVOList.size());

Conversation

lhpqaq commented Mar 26, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

Pull request overview

Reviewed changes

Uh oh!

Copilot AI Mar 29, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Mar 29, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Mar 29, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Mar 29, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Mar 29, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Mar 29, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Mar 29, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Mar 29, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Mar 29, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Mar 29, 2026

Choose a reason for hiding this comment

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants

lhpqaq commented Mar 26, 2026 •

edited

Loading