@theia/ai-chat 1.56.0 → 1.57.0-next.136

package/src/common/orchestrator-chat-agent.ts

@@ -1,179 +0,0 @@
-// *****************************************************************************
-// Copyright (C) 2024 EclipseSource GmbH.
-//
-// This program and the accompanying materials are made available under the
-// terms of the Eclipse Public License v. 2.0 which is available at
-// http://www.eclipse.org/legal/epl-2.0.
-//
-// This Source Code may also be made available under the following Secondary
-// Licenses when the conditions for such availability set forth in the Eclipse
-// Public License v. 2.0 are satisfied: GNU General Public License, version 2
-// with the GNU Classpath Exception which is available at
-// https://www.gnu.org/software/classpath/license.html.
-//
-// SPDX-License-Identifier: EPL-2.0 OR GPL-2.0-only WITH Classpath-exception-2.0
-// *****************************************************************************
-
-import { AgentSpecificVariables, getJsonOfText, getTextOfResponse, LanguageModelResponse } from '@theia/ai-core';
-import {
-    PromptTemplate
-} from '@theia/ai-core/lib/common';
-import { inject, injectable } from '@theia/core/shared/inversify';
-import { ChatAgentService } from './chat-agent-service';
-import { AbstractStreamParsingChatAgent, ChatAgent, SystemMessageDescription } from './chat-agents';
-import { ChatRequestModelImpl, InformationalChatResponseContentImpl } from './chat-model';
-import { generateUuid } from '@theia/core';
-import { ChatHistoryEntry } from './chat-history-entry';
-
-export const orchestratorTemplate: PromptTemplate = {
-    id: 'orchestrator-system',
-    template: `{{!-- Made improvements or adaptations to this prompt template? We’d love for you to share it with the community! Contribute back here:
-https://github.com/eclipse-theia/theia/discussions/new?category=prompt-template-contribution --}}
-# Instructions
-
-Your task is to identify which Chat Agent(s) should best reply a given user's message.
-You consider all messages of the conversation to ensure consistency and avoid agent switches without a clear context change.
-You should select the best Chat Agent based on the name and description of the agents, matching them to the user message.
-
-## Constraints
-
-Your response must be a JSON array containing the id(s) of the selected Chat Agent(s).
-
-* Do not use ids that are not provided in the list below.
-* Do not include any additional information, explanations, or questions for the user.
-* If there is no suitable choice, pick \`Universal\`.
-* If there are multiple good choices, return all of them.
-
-Unless there is a more specific agent available, select \`Universal\`, especially for general programming-related questions.
-You must only use the \`id\` attribute of the agent, never the name.
-
-### Example Results
-
-\`\`\`json
-["Universal"]
-\`\`\`
-
-\`\`\`json
-["AnotherChatAgent", "Universal"]
-\`\`\`
-
-## List of Currently Available Chat Agents
-
-{{chatAgents}}
-`};
-
-export const OrchestratorChatAgentId = 'Orchestrator';
-const OrchestratorRequestIdKey = 'orchestatorRequestIdKey';
-
-@injectable()
-export class OrchestratorChatAgent extends AbstractStreamParsingChatAgent implements ChatAgent {
-    name: string;
-    description: string;
-    readonly variables: string[];
-    promptTemplates: PromptTemplate[];
-    fallBackChatAgentId: string;
-    readonly functions: string[] = [];
-    readonly agentSpecificVariables: AgentSpecificVariables[] = [];
-
-    constructor() {
-        super(OrchestratorChatAgentId, [{
-            purpose: 'agent-selection',
-            identifier: 'openai/gpt-4o',
-        }], 'agent-selection', 'codicon codicon-symbol-boolean', undefined, undefined, false);
-        this.name = OrchestratorChatAgentId;
-        this.description = 'This agent analyzes the user request against the description of all available chat agents and selects the best fitting agent to answer the request \
-        (by using AI).The user\'s request will be directly delegated to the selected agent without further confirmation.';
-        this.variables = ['chatAgents'];
-        this.promptTemplates = [orchestratorTemplate];
-        this.fallBackChatAgentId = 'Universal';
-        this.functions = [];
-        this.agentSpecificVariables = [];
-    }
-
-    @inject(ChatAgentService)
-    protected chatAgentService: ChatAgentService;
-
-    override async invoke(request: ChatRequestModelImpl): Promise<void> {
-        request.response.addProgressMessage({ content: 'Determining the most appropriate agent', status: 'inProgress' });
-        // We generate a dedicated ID for recording the orchestrator request/response, as we will forward the original request to another agent
-        const orchestratorRequestId = generateUuid();
-        request.addData(OrchestratorRequestIdKey, orchestratorRequestId);
-        const messages = await this.getMessages(request.session);
-        const systemMessage = (await this.getSystemMessageDescription())?.text;
-        this.recordingService.recordRequest(
-            ChatHistoryEntry.fromRequest(this.id, request, {
-                requestId: orchestratorRequestId,
-                messages,
-                systemMessage
-            })
-        );
-        return super.invoke(request);
-    }
-
-    protected async getSystemMessageDescription(): Promise<SystemMessageDescription | undefined> {
-        const resolvedPrompt = await this.promptService.getPrompt(orchestratorTemplate.id);
-        return resolvedPrompt ? SystemMessageDescription.fromResolvedPromptTemplate(resolvedPrompt) : undefined;
-    }
-
-    protected override async addContentsToResponse(response: LanguageModelResponse, request: ChatRequestModelImpl): Promise<void> {
-        let agentIds: string[] = [];
-        const responseText = await getTextOfResponse(response);
-        // We use the previously generated, dedicated ID to log the orchestrator response before we forward the original request
-        const orchestratorRequestId = request.getDataByKey(OrchestratorRequestIdKey);
-        if (typeof orchestratorRequestId === 'string') {
-            this.recordingService.recordResponse({
-                agentId: this.id,
-                sessionId: request.session.id,
-                requestId: orchestratorRequestId,
-                response: responseText,
-            });
-        }
-        try {
-            const jsonResponse = await getJsonOfText(responseText);
-            if (Array.isArray(jsonResponse)) {
-                agentIds = jsonResponse.filter((id: string) => id !== this.id);
-            }
-        } catch (error: unknown) {
-            // The llm sometimes does not return a parseable result
-            this.logger.error('Failed to parse JSON response', error);
-        }
-
-        if (agentIds.length < 1) {
-            this.logger.error('No agent was selected, delegating to fallback chat agent');
-            request.response.progressMessages.forEach(progressMessage =>
-                request.response.updateProgressMessage({ ...progressMessage, status: 'failed' })
-            );
-            agentIds = [this.fallBackChatAgentId];
-        }
-
-        // check if selected (or fallback) agent exists
-        if (!this.chatAgentService.getAgent(agentIds[0])) {
-            this.logger.error(`Chat agent ${agentIds[0]} not found. Falling back to first registered agent.`);
-            const firstRegisteredAgent = this.chatAgentService.getAgents().filter(a => a.id !== this.id)[0]?.id;
-            if (firstRegisteredAgent) {
-                agentIds = [firstRegisteredAgent];
-            } else {
-                throw new Error('No chat agent available to handle request. Please check your configuration whether any are enabled.');
-            }
-        }
-
-        // TODO support delegating to more than one agent
-        const delegatedToAgent = agentIds[0];
-        request.response.response.addContent(new InformationalChatResponseContentImpl(
-            `*Orchestrator*: Delegating to \`@${delegatedToAgent}\`
-            
-            ---
-
-            `
-        ));
-        request.response.overrideAgentId(delegatedToAgent);
-        request.response.progressMessages.forEach(progressMessage =>
-            request.response.updateProgressMessage({ ...progressMessage, status: 'completed' })
-        );
-        const agent = this.chatAgentService.getAgent(delegatedToAgent);
-        if (!agent) {
-            throw new Error(`Chat agent ${delegatedToAgent} not found.`);
-        }
-        await agent.invoke(request);
-    }
-}

package/src/common/universal-chat-agent.ts

@@ -1,117 +0,0 @@
-// *****************************************************************************
-// Copyright (C) 2024 EclipseSource GmbH.
-//
-// This program and the accompanying materials are made available under the
-// terms of the Eclipse Public License v. 2.0 which is available at
-// http://www.eclipse.org/legal/epl-2.0.
-//
-// This Source Code may also be made available under the following Secondary
-// Licenses when the conditions for such availability set forth in the Eclipse
-// Public License v. 2.0 are satisfied: GNU General Public License, version 2
-// with the GNU Classpath Exception which is available at
-// https://www.gnu.org/software/classpath/license.html.
-//
-// SPDX-License-Identifier: EPL-2.0 OR GPL-2.0-only WITH Classpath-exception-2.0
-// *****************************************************************************
-
-import { AgentSpecificVariables } from '@theia/ai-core';
-import {
-   PromptTemplate
-} from '@theia/ai-core/lib/common';
-import { injectable } from '@theia/core/shared/inversify';
-import { AbstractStreamParsingChatAgent, ChatAgent, SystemMessageDescription } from './chat-agents';
-
-export const universalTemplate: PromptTemplate = {
-   id: 'universal-system',
-   template: `{{!-- Made improvements or adaptations to this prompt template? We’d love for you to share it with the community! Contribute back here:
-https://github.com/eclipse-theia/theia/discussions/new?category=prompt-template-contribution --}}
-# Instructions
-
-You are an AI assistant integrated into the Theia IDE, specifically designed to help software developers by
-providing concise and accurate answers to programming-related questions. Your role is to enhance the
-developer's productivity by offering quick solutions, explanations, and best practices.
-Keep responses short and to the point, focusing on delivering valuable insights, best practices and
-simple solutions.
-
-### Guidelines
-
-1. **Understand Context:**
-   - Assess the context of the code or issue when available.
-   - Tailor responses to be relevant to the programming language, framework, or tools like Eclipse Theia.
-   - Ask clarifying questions if necessary to provide accurate assistance.
-
-2. **Provide Clear Solutions:**
-   - Offer direct answers or code snippets that solve the problem or clarify the concept.
-   - Avoid lengthy explanations unless necessary for understanding.
-
-3. **Promote Best Practices:**
-   - Suggest best practices and common patterns relevant to the question.
-   - Provide links to official documentation for further reading when applicable.
-
-4. **Support Multiple Languages and Tools:**
-   - Be familiar with popular programming languages, frameworks, IDEs like Eclipse Theia, and command-line tools.
-   - Adapt advice based on the language, environment, or tools specified by the developer.
-
-5. **Facilitate Learning:**
-   - Encourage learning by explaining why a solution works or why a particular approach is recommended.
-   - Keep explanations concise and educational.
-
-6. **Maintain Professional Tone:**
-   - Communicate in a friendly, professional manner.
-   - Use technical jargon appropriately, ensuring clarity for the target audience.
-
-7. **Stay on Topic:**
-   - Limit responses strictly to topics related to software development, frameworks, Eclipse Theia, terminal usage, and relevant technologies.
-   - Politely decline to answer questions unrelated to these areas by saying, "I'm here to assist with programming-related questions.
-     For other topics, please refer to a specialized source."
-
-### Example Interactions
-
-- **Question:** "What's the difference between \`let\` and \`var\` in JavaScript?"
-  **Answer:** "\`let\` is block-scoped, while \`var\` is function-scoped. Prefer \`let\` to avoid scope-related bugs."
-
-- **Question:** "How do I handle exceptions in Java?"
-  **Answer:** "Use try-catch blocks: \`\`\`java try { /* code */ } catch (ExceptionType e) { /* handle exception */ }\`\`\`."
-
-- **Question:** "What is the capital of France?"
-  **Answer:** "I'm here to assist with programming-related queries. For other topics, please refer to a specialized source."
-`
-};
-
-export const universalTemplateVariant: PromptTemplate = {
-   id: 'universal-system-empty',
-   template: '',
-   variantOf: universalTemplate.id,
-};
-
-@injectable()
-export class UniversalChatAgent extends AbstractStreamParsingChatAgent implements ChatAgent {
-   name: string;
-   description: string;
-   variables: string[];
-   promptTemplates: PromptTemplate[];
-   readonly functions: string[];
-   readonly agentSpecificVariables: AgentSpecificVariables[];
-
-   constructor() {
-      super('Universal', [{
-         purpose: 'chat',
-         identifier: 'openai/gpt-4o',
-      }], 'chat');
-      this.name = 'Universal';
-      this.description = 'This agent is designed to help software developers by providing concise and accurate '
-         + 'answers to general programming and software development questions. It is also the fall-back for any generic '
-         + 'questions the user might ask. The universal agent currently does not have any context by default, i.e. it cannot '
-         + 'access the current user context or the workspace.';
-      this.variables = [];
-      this.promptTemplates = [universalTemplate, universalTemplateVariant];
-      this.functions = [];
-      this.agentSpecificVariables = [];
-   }
-
-   protected override async getSystemMessageDescription(): Promise<SystemMessageDescription | undefined> {
-      const resolvedPrompt = await this.promptService.getPrompt(universalTemplate.id);
-      return resolvedPrompt ? SystemMessageDescription.fromResolvedPromptTemplate(resolvedPrompt) : undefined;
-   }
-
-}