Skip to content

An AI-powered task-management system you can drop into Cursor, Lovable, Windsurf, Roo, and others.

Adjusts claude calls using message to use stream instead.

Eyal Toledano341fbdf40fce
diff --git a/package.json b/package.json
index 152fda6d..7b18f1dd 100644
--- a/package.json
+++ b/package.json
@@ -1,6 +1,6 @@
{
"name": "task-master-ai",
- "version": "0.9.19",
+ "version": "0.9.22",
"description": "A task management system for ambitious AI-driven development that doesn't overwhelm and confuse Cursor.",
"main": "index.js",
"type": "module",
diff --git a/scripts/modules/ai-services.js b/scripts/modules/ai-services.js
index 0128ad6f..cc3e8169 100644
--- a/scripts/modules/ai-services.js
+++ b/scripts/modules/ai-services.js
@@ -123,9 +123,11 @@ Important: Your response must be valid JSON only, with no additional explanation
async function handleStreamingRequest(prdContent, prdPath, numTasks, maxTokens, systemPrompt) {
const loadingIndicator = startLoadingIndicator('Generating tasks from PRD...');
let responseText = '';
+ let streamingInterval = null;
try {
- const message = await anthropic.messages.create({
+ // Use streaming for handling large responses
+ const stream = await anthropic.messages.create({
model: CONFIG.model,
max_tokens: maxTokens,
temperature: CONFIG.temperature,
@@ -135,14 +137,34 @@ async function handleStreamingRequest(prdContent, prdPath, numTasks, maxTokens,
role: 'user',
content: `Here's the Product Requirements Document (PRD) to break down into ${numTasks} tasks:\n\n${prdContent}`
}
- ]
+ ],
+ stream: true
});
- responseText = message.content[0].text;
+ // Update loading indicator to show streaming progress
+ let dotCount = 0;
+ const readline = await import('readline');
+ streamingInterval = setInterval(() => {
+ readline.cursorTo(process.stdout, 0);
+ process.stdout.write(`Receiving streaming response from Claude${'.'.repeat(dotCount)}`);
+ dotCount = (dotCount + 1) % 4;
+ }, 500);
+
+ // Process the stream
+ for await (const chunk of stream) {
+ if (chunk.type === 'content_block_delta' && chunk.delta.text) {
+ responseText += chunk.delta.text;
+ }
+ }
+
+ if (streamingInterval) clearInterval(streamingInterval);
stopLoadingIndicator(loadingIndicator);
+ log('info', "Completed streaming response from Claude API!");
+
return processClaudeResponse(responseText, numTasks, 0, prdContent, prdPath);
} catch (error) {
+ if (streamingInterval) clearInterval(streamingInterval);
stopLoadingIndicator(loadingIndicator);
throw error;
}
@@ -224,6 +246,8 @@ async function generateSubtasks(task, numSubtasks, nextSubtaskId, additionalCont
log('info', `Generating ${numSubtasks} subtasks for task ${task.id}: ${task.title}`);
const loadingIndicator = startLoadingIndicator(`Generating subtasks for task ${task.id}...`);
+ let streamingInterval = null;
+ let responseText = '';
const systemPrompt = `You are an AI assistant helping with task breakdown for software development.
You need to break down a high-level task into ${numSubtasks} specific subtasks that can be implemented one by one.
@@ -269,22 +293,49 @@ Return exactly ${numSubtasks} subtasks with the following JSON structure:
Note on dependencies: Subtasks can depend on other subtasks with lower IDs. Use an empty array if there are no dependencies.`;
- const message = await anthropic.messages.create({
- model: CONFIG.model,
- max_tokens: CONFIG.maxTokens,
- temperature: CONFIG.temperature,
- system: systemPrompt,
- messages: [
- {
- role: 'user',
- content: userPrompt
+ try {
+ // Update loading indicator to show streaming progress
+ let dotCount = 0;
+ const readline = await import('readline');
+ streamingInterval = setInterval(() => {
+ readline.cursorTo(process.stdout, 0);
+ process.stdout.write(`Generating subtasks for task ${task.id}${'.'.repeat(dotCount)}`);
+ dotCount = (dotCount + 1) % 4;
+ }, 500);
+
+ // Use streaming API call
+ const stream = await anthropic.messages.create({
+ model: CONFIG.model,
+ max_tokens: CONFIG.maxTokens,
+ temperature: CONFIG.temperature,
+ system: systemPrompt,
+ messages: [
+ {
+ role: 'user',
+ content: userPrompt
+ }
+ ],
+ stream: true
+ });
+
+ // Process the stream
+ for await (const chunk of stream) {
+ if (chunk.type === 'content_block_delta' && chunk.delta.text) {
+ responseText += chunk.delta.text;
}
- ]
- });
-
- stopLoadingIndicator(loadingIndicator);
-
- return parseSubtasksFromText(message.content[0].text, nextSubtaskId, numSubtasks, task.id);
+ }
+
+ if (streamingInterval) clearInterval(streamingInterval);
+ stopLoadingIndicator(loadingIndicator);
+
+ log('info', `Completed generating subtasks for task ${task.id}`);
+
+ return parseSubtasksFromText(responseText, nextSubtaskId, numSubtasks, task.id);
+ } catch (error) {
+ if (streamingInterval) clearInterval(streamingInterval);
+ stopLoadingIndicator(loadingIndicator);
+ throw error;
+ }
} catch (error) {
log('error', `Error generating subtasks: ${error.message}`);
throw error;
@@ -339,6 +390,8 @@ ${additionalContext || "No additional context provided."}
// Now generate subtasks with Claude
const loadingIndicator = startLoadingIndicator(`Generating research-backed subtasks for task ${task.id}...`);
+ let streamingInterval = null;
+ let responseText = '';
const systemPrompt = `You are an AI assistant helping with task breakdown for software development.
You need to break down a high-level task into ${numSubtasks} specific subtasks that can be implemented one by one.
@@ -350,7 +403,7 @@ Subtasks should:
1. Be specific and actionable implementation steps
2. Follow a logical sequence
3. Each handle a distinct part of the parent task
-4. Include clear guidance on implementation approach, referencing the research where relevant
+4. Include clear guidance on implementation approach
5. Have appropriate dependency chains between subtasks
6. Collectively cover all aspects of the parent task
@@ -362,8 +415,7 @@ For each subtask, provide:
Each subtask should be implementable in a focused coding session.`;
- const userPrompt = `Please break down this task into ${numSubtasks} specific, actionable subtasks,
-using the research findings to inform your breakdown:
+ const userPrompt = `Please break down this task into ${numSubtasks} specific, well-researched, actionable subtasks:
Task ID: ${task.id}
Title: ${task.title}
@@ -377,31 +429,58 @@ Return exactly ${numSubtasks} subtasks with the following JSON structure:
{
"id": ${nextSubtaskId},
"title": "First subtask title",
- "description": "Detailed description",
+ "description": "Detailed description incorporating research",
"dependencies": [],
- "details": "Implementation details"
+ "details": "Implementation details with best practices"
},
...more subtasks...
]
Note on dependencies: Subtasks can depend on other subtasks with lower IDs. Use an empty array if there are no dependencies.`;
- const message = await anthropic.messages.create({
- model: CONFIG.model,
- max_tokens: CONFIG.maxTokens,
- temperature: CONFIG.temperature,
- system: systemPrompt,
- messages: [
- {
- role: 'user',
- content: userPrompt
+ try {
+ // Update loading indicator to show streaming progress
+ let dotCount = 0;
+ const readline = await import('readline');
+ streamingInterval = setInterval(() => {
+ readline.cursorTo(process.stdout, 0);
+ process.stdout.write(`Generating research-backed subtasks for task ${task.id}${'.'.repeat(dotCount)}`);
+ dotCount = (dotCount + 1) % 4;
+ }, 500);
+
+ // Use streaming API call
+ const stream = await anthropic.messages.create({
+ model: CONFIG.model,
+ max_tokens: CONFIG.maxTokens,
+ temperature: CONFIG.temperature,
+ system: systemPrompt,
+ messages: [
+ {
+ role: 'user',
+ content: userPrompt
+ }
+ ],
+ stream: true
+ });
+
+ // Process the stream
+ for await (const chunk of stream) {
+ if (chunk.type === 'content_block_delta' && chunk.delta.text) {
+ responseText += chunk.delta.text;
}
- ]
- });
-
- stopLoadingIndicator(loadingIndicator);
-
- return parseSubtasksFromText(message.content[0].text, nextSubtaskId, numSubtasks, task.id);
+ }
+
+ if (streamingInterval) clearInterval(streamingInterval);
+ stopLoadingIndicator(loadingIndicator);
+
+ log('info', `Completed generating research-backed subtasks for task ${task.id}`);
+
+ return parseSubtasksFromText(responseText, nextSubtaskId, numSubtasks, task.id);
+ } catch (error) {
+ if (streamingInterval) clearInterval(streamingInterval);
+ stopLoadingIndicator(loadingIndicator);
+ throw error;
+ }
} catch (error) {
log('error', `Error generating research-backed subtasks: ${error.message}`);
throw error;
diff --git a/scripts/modules/task-manager.js b/scripts/modules/task-manager.js
index 3cc96220..ad14d11a 100644
--- a/scripts/modules/task-manager.js
+++ b/scripts/modules/task-manager.js
@@ -243,38 +243,65 @@ Return only the updated tasks as a valid JSON array.`
const jsonText = responseText.substring(jsonStart, jsonEnd + 1);
updatedTasks = JSON.parse(jsonText);
} else {
- // Call Claude to update the tasks
- const message = await anthropic.messages.create({
- model: CONFIG.model,
- max_tokens: CONFIG.maxTokens,
- temperature: CONFIG.temperature,
- system: systemPrompt,
- messages: [
- {
- role: 'user',
- content: `Here are the tasks to update:
+ // Call Claude to update the tasks with streaming enabled
+ let responseText = '';
+ let streamingInterval = null;
+
+ try {
+ // Update loading indicator to show streaming progress
+ let dotCount = 0;
+ const readline = await import('readline');
+ streamingInterval = setInterval(() => {
+ readline.cursorTo(process.stdout, 0);
+ process.stdout.write(`Receiving streaming response from Claude${'.'.repeat(dotCount)}`);
+ dotCount = (dotCount + 1) % 4;
+ }, 500);
+
+ // Use streaming API call
+ const stream = await anthropic.messages.create({
+ model: CONFIG.model,
+ max_tokens: CONFIG.maxTokens,
+ temperature: CONFIG.temperature,
+ system: systemPrompt,
+ messages: [
+ {
+ role: 'user',
+ content: `Here are the tasks to update:
${taskData}
Please update these tasks based on the following new context:
${prompt}
Return only the updated tasks as a valid JSON array.`
+ }
+ ],
+ stream: true
+ });
+
+ // Process the stream
+ for await (const chunk of stream) {
+ if (chunk.type === 'content_block_delta' && chunk.delta.text) {
+ responseText += chunk.delta.text;
}
- ]
- });
-
- const responseText = message.content[0].text;
-
- // Extract JSON from response
- const jsonStart = responseText.indexOf('[');
- const jsonEnd = responseText.lastIndexOf(']');
-
- if (jsonStart === -1 || jsonEnd === -1) {
- throw new Error("Could not find valid JSON array in Claude's response");
+ }
+
+ if (streamingInterval) clearInterval(streamingInterval);
+ log('info', "Completed streaming response from Claude API!");
+
+ // Extract JSON from response
+ const jsonStart = responseText.indexOf('[');
+ const jsonEnd = responseText.lastIndexOf(']');
+
+ if (jsonStart === -1 || jsonEnd === -1) {
+ throw new Error("Could not find valid JSON array in Claude's response");
+ }
+
+ const jsonText = responseText.substring(jsonStart, jsonEnd + 1);
+ updatedTasks = JSON.parse(jsonText);
+ } catch (error) {
+ if (streamingInterval) clearInterval(streamingInterval);
+ throw error;
}
-
- const jsonText = responseText.substring(jsonStart, jsonEnd + 1);
- updatedTasks = JSON.parse(jsonText);
}
// Replace the tasks in the original data