claude-flow
Version:
Ruflo - Enterprise AI agent orchestration for Claude Code. Deploy 60+ specialized agents in coordinated swarms with self-learning, fault-tolerant consensus, vector memory, and MCP integration
238 lines (213 loc) • 8.92 kB
text/typescript
import "https://deno.land/x/xhr@0.1.0/mod.ts";
import { serve } from "https://deno.land/std@0.168.0/http/server.ts";
const corsHeaders = {
'Access-Control-Allow-Origin': '*',
'Access-Control-Allow-Headers': 'authorization, x-client-info, apikey, content-type',
};
interface ActionItemsRequest {
goal: string;
researchContext: Array<{
stepTitle: string;
findings: Array<{
title: string;
content: string;
source?: string;
}>;
}>;
totalSteps: number;
totalDataPoints: number;
}
serve(async (req) => {
if (req.method === 'OPTIONS') {
return new Response(null, { headers: corsHeaders });
}
try {
const { goal, researchContext, totalSteps, totalDataPoints }: ActionItemsRequest = await req.json();
console.log('Generating action items for goal:', goal);
const LOVABLE_API_KEY = Deno.env.get('LOVABLE_API_KEY');
if (!LOVABLE_API_KEY) {
throw new Error('LOVABLE_API_KEY is not configured');
}
// Build research summary from all steps
let researchSummary = '';
researchContext.forEach(step => {
researchSummary += `\n${step.stepTitle}:\n`;
step.findings.forEach(finding => {
researchSummary += `• ${finding.title}: ${finding.content}\n`;
if (finding.source) researchSummary += ` Source: ${finding.source}\n`;
});
});
const systemPrompt = `You are an expert strategic planner and implementation consultant. Generate contextual, actionable recommendations based on research findings.
CRITICAL INSTRUCTIONS:
- Generate action items that are DIRECTLY RELEVANT to the research goal
- Base recommendations on ACTUAL research findings provided
- Do NOT use generic "pilot program" or "scale to production" templates unless they make sense for this specific goal
- Tailor action items to the domain and context of the research
- Include specific, actionable steps with realistic timelines and resources
For example:
- If researching "best family car" → recommend specific car models, comparison steps, test drives
- If researching "law school alternatives" → recommend specific programs, application steps, bar exam prep
- If researching "quantum computing" → recommend learning paths, tools, research papers
- If researching business strategies → recommend market analysis, competitor research, implementation plans`;
const userPrompt = `
RESEARCH GOAL: ${goal}
RESEARCH FINDINGS (${totalSteps} steps, ${totalDataPoints} data points):
${researchSummary}
Generate 3-4 CONTEXTUAL action items that directly help achieve or implement the research goal based on these findings.
REQUIREMENTS:
1. Each action item must be SPECIFIC to "${goal}" - not generic project management steps
2. Reference actual research findings in the description
3. Provide realistic timelines appropriate for the goal (not always "Week 1-4")
4. Include relevant resources and metrics for this specific domain
5. Identify domain-specific risks and mitigation strategies
Also generate a comprehensive 2-3 paragraph executive summary that:
- Directly addresses what was learned about "${goal}"
- Highlights the most important findings with specifics
- Provides clear conclusions and recommendations based on the research
Format:
{
"actionItems": [
{
"id": "1",
"title": "Specific action relevant to ${goal}",
"description": "Detailed description referencing actual research findings...",
"timeline": "Appropriate timeline (e.g., '1-2 weeks', '3 months', 'Immediately')",
"timelineDetails": "Breakdown of timeline phases",
"priority": "High" | "Medium" | "Low",
"resources": {
"budget": "Realistic budget if applicable, or 'Minimal cost' or 'Research only'",
"team": "Required people/roles",
"tools": ["Domain-specific tools/resources"]
},
"metrics": ["Specific success metrics for this action"],
"risks": [
{
"risk": "Domain-specific risk",
"mitigation": "Realistic mitigation strategy"
}
],
"references": [
{ "title": "Relevant resource", "url": "URL if applicable" }
],
"researchContext": "How this connects to research findings"
}
],
"summary": "Comprehensive 2-3 paragraph executive summary addressing the research goal with specific findings and recommendations..."
}`;
const response = await fetch('https://ai.gateway.lovable.dev/v1/chat/completions', {
method: 'POST',
headers: {
'Authorization': `Bearer ${LOVABLE_API_KEY}`,
'Content-Type': 'application/json',
},
body: JSON.stringify({
model: 'google/gemini-2.5-flash',
messages: [
{ role: 'system', content: systemPrompt },
{ role: 'user', content: userPrompt }
],
tools: [
{
type: "function",
function: {
name: "generate_action_plan",
description: "Generate contextual action items and executive summary based on research findings",
parameters: {
type: "object",
properties: {
actionItems: {
type: "array",
items: {
type: "object",
properties: {
id: { type: "string" },
title: { type: "string" },
description: { type: "string" },
timeline: { type: "string" },
timelineDetails: { type: "string" },
priority: { type: "string", enum: ["High", "Medium", "Low"] },
resources: {
type: "object",
properties: {
budget: { type: "string" },
team: { type: "string" },
tools: { type: "array", items: { type: "string" } }
}
},
metrics: { type: "array", items: { type: "string" } },
risks: {
type: "array",
items: {
type: "object",
properties: {
risk: { type: "string" },
mitigation: { type: "string" }
}
}
},
references: {
type: "array",
items: {
type: "object",
properties: {
title: { type: "string" },
url: { type: "string" }
}
}
},
researchContext: { type: "string" }
},
required: ["id", "title", "description", "timeline", "priority", "resources", "metrics"]
}
},
summary: {
type: "string",
description: "Comprehensive executive summary (2-3 paragraphs)"
}
},
required: ["actionItems", "summary"]
}
}
}
],
tool_choice: { type: "function", function: { name: "generate_action_plan" } }
}),
});
if (!response.ok) {
if (response.status === 429) {
return new Response(JSON.stringify({ error: "Rate limits exceeded" }), {
status: 429,
headers: { ...corsHeaders, 'Content-Type': 'application/json' },
});
}
if (response.status === 402) {
return new Response(JSON.stringify({ error: "AI usage limit reached" }), {
status: 402,
headers: { ...corsHeaders, 'Content-Type': 'application/json' },
});
}
throw new Error(`AI gateway error: ${response.status}`);
}
const data = await response.json();
const toolCall = data.choices?.[0]?.message?.tool_calls?.[0];
if (!toolCall) {
throw new Error('No tool call in AI response');
}
const result = JSON.parse(toolCall.function.arguments);
console.log('Generated action items:', result.actionItems?.length || 0);
return new Response(JSON.stringify(result), {
headers: { ...corsHeaders, 'Content-Type': 'application/json' },
});
} catch (error) {
console.error('Error in generate-action-items function:', error);
return new Response(
JSON.stringify({
error: error instanceof Error ? error.message : 'Unknown error occurred'
}),
{
status: 500,
headers: { ...corsHeaders, 'Content-Type': 'application/json' },
}
);
}
});