Spaces:
Sleeping
Sleeping
File size: 15,769 Bytes
fd1a435 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 | # extractor_agent_runner.py
import asyncio
import json
from typing import Dict, Any, Optional
from datetime import datetime
from openai import AsyncOpenAI
from pydantic import BaseModel, Field
from firebase_admin import db
# ------------------------
# Pydantic Models for Structured Outputs
# ------------------------
class BusinessExtraction(BaseModel):
"""Extracted business information"""
business_name: str = Field(description="Name of the business idea")
industry: str = Field(description="Industry or sector")
target_audience: str = Field(description="Target customer base")
core_problem: str = Field(description="Main problem being solved")
solution: str = Field(description="Proposed solution")
key_features: list[str] = Field(description="List of key features or capabilities")
tech_stack: list[str] = Field(description="Recommended technology stack")
estimated_complexity: str = Field(description="Low, Medium, or High")
class Config:
extra = "forbid"
class AgentSpecification(BaseModel):
"""AI Agent specifications"""
agent_name: str = Field(description="Name of the AI agent")
agent_purpose: str = Field(description="Main purpose of the agent")
capabilities: list[str] = Field(description="List of agent capabilities")
integrations: list[str] = Field(description="Required integrations or APIs")
data_requirements: list[str] = Field(description="Data needed for the agent")
deployment_model: str = Field(description="e.g., Cloud, On-premise, Hybrid")
class Config:
extra = "forbid"
class Phase(BaseModel):
"""Single implementation phase"""
phase: str = Field(description="Phase name")
duration: str = Field(description="Time duration")
tasks: list[str] = Field(description="List of tasks in this phase")
class Config:
extra = "forbid"
class ImplementationPlan(BaseModel):
"""Implementation roadmap"""
phases: list[Phase] = Field(description="Implementation phases with timeline")
estimated_timeline: str = Field(description="Overall timeline (e.g., 3-6 months)")
team_requirements: list[str] = Field(description="Required team members and roles")
estimated_cost: str = Field(description="Estimated cost range")
risks: list[str] = Field(description="Potential risks and challenges")
success_metrics: list[str] = Field(description="KPIs to measure success")
class Config:
extra = "forbid"
class CompleteExtraction(BaseModel):
"""Complete extraction result"""
business: BusinessExtraction
agent: AgentSpecification
implementation: ImplementationPlan
summary: str = Field(description="Executive summary of the entire plan")
class Config:
extra = "forbid"
# ------------------------
# Agent Orchestrator
# ------------------------
class AgentOrchestrator:
"""
Orchestrates multiple AI agents for business idea extraction.
Supports OpenAI GPT, Google Gemini, and xAI Grok models.
"""
def __init__(self, session_id: str):
self.session_id = session_id
self.client: Optional[AsyncOpenAI] = None
self.model: str = ""
async def _emit_log(self, message: str, type: str = "log"):
"""Emit log to Firebase"""
try:
ref = db.reference(f'sessions/{self.session_id}/logs')
ref.push({
'message': message,
'type': type,
'timestamp': datetime.now().isoformat()
})
except Exception as e:
print(f"Error emitting log: {e}")
def _setup_client(self, model: str, api_key: str):
"""Setup appropriate client based on model type"""
model_lower = model.lower()
if "gpt" in model_lower or "o1" in model_lower:
# OpenAI models
self.client = AsyncOpenAI(api_key=api_key)
self.model = model
elif "grok" in model_lower:
# xAI Grok models (OpenAI-compatible)
self.client = AsyncOpenAI(
api_key=api_key,
base_url="https://api.x.ai/v1"
)
self.model = model
elif "gemini" in model_lower:
# Google Gemini - use direct API
self.client = None # Will handle separately
self.model = model
else:
raise ValueError(f"Unsupported model: {model}")
async def _call_openai_structured(self, system_prompt: str, user_prompt: str, response_format: type[BaseModel]) -> BaseModel:
"""Call OpenAI with structured output"""
try:
response = await self.client.beta.chat.completions.parse(
model=self.model,
messages=[
{"role": "system", "content": system_prompt},
{"role": "user", "content": user_prompt}
],
response_format=response_format,
temperature=0.7
)
return response.choices[0].message.parsed
except Exception as e:
raise Exception(f"OpenAI API call failed: {str(e)}")
async def _call_gemini_structured(self, system_prompt: str, user_prompt: str, api_key: str, response_format: type[BaseModel]) -> BaseModel:
"""Call Gemini and parse JSON response into Pydantic model"""
try:
import google.generativeai as genai
genai.configure(api_key=api_key)
model = genai.GenerativeModel(self.model)
# Get schema for structured output
schema_example = response_format.model_json_schema()
full_prompt = f"""{system_prompt}
User Query: {user_prompt}
Respond ONLY with valid JSON matching this schema:
{json.dumps(schema_example, indent=2)}
Important: No markdown, no backticks, just pure JSON."""
response = await model.generate_content_async(full_prompt)
# Extract JSON from response
text = response.text.strip()
if text.startswith("```json"):
text = text[7:]
if text.startswith("```"):
text = text[3:]
if text.endswith("```"):
text = text[:-3]
text = text.strip()
# Parse and validate with Pydantic
data = json.loads(text)
return response_format(**data)
except Exception as e:
raise Exception(f"Gemini API call failed: {str(e)}")
async def _extract_business_info(self, query: str, api_key: str) -> BusinessExtraction:
"""Extract business information from query"""
await self._emit_log("π Analyzing business idea...", "system")
system_prompt = """You are a business analyst expert. Extract detailed business information from the user's query.
Be thorough and specific. If information is not explicitly stated, make reasonable inferences based on the context."""
user_prompt = f"""Analyze this business idea and extract key information:
Query: {query}
Provide detailed extraction including:
- Business name (create a suitable name if not provided)
- Industry classification
- Target audience
- Core problem being solved
- Proposed solution
- Key features (at least 3-5)
- Recommended tech stack
- Estimated complexity (Low/Medium/High)"""
if "gemini" in self.model.lower():
# Gemini: Parse JSON manually
return await self._call_gemini_structured(system_prompt, user_prompt, api_key, BusinessExtraction)
else:
# OpenAI/Grok: Use structured outputs
return await self._call_openai_structured(system_prompt, user_prompt, BusinessExtraction)
async def _extract_agent_specs(self, query: str, business: BusinessExtraction, api_key: str) -> AgentSpecification:
"""Extract AI agent specifications"""
await self._emit_log("π€ Designing AI agent specifications...", "system")
system_prompt = """You are an AI agent architect. Design detailed specifications for an AI agent based on the business requirements.
Be specific about capabilities, integrations, and technical requirements."""
user_prompt = f"""Design an AI agent for this business:
Business: {business.business_name}
Industry: {business.industry}
Problem: {business.core_problem}
Solution: {business.solution}
Create detailed agent specifications including:
- Agent name
- Primary purpose
- Specific capabilities (at least 5)
- Required integrations (APIs, databases, services)
- Data requirements
- Deployment model (Cloud/On-premise/Hybrid)"""
if "gemini" in self.model.lower():
return await self._call_gemini_structured(system_prompt, user_prompt, api_key, AgentSpecification)
else:
return await self._call_openai_structured(system_prompt, user_prompt, AgentSpecification)
async def _create_implementation_plan(self, query: str, business: BusinessExtraction, agent: AgentSpecification, api_key: str) -> ImplementationPlan:
"""Create implementation roadmap"""
await self._emit_log("π Creating implementation roadmap...", "system")
system_prompt = """You are a project manager and implementation strategist. Create a detailed implementation plan.
Include realistic timelines, resource requirements, and risk assessments."""
user_prompt = f"""Create an implementation plan for:
Business: {business.business_name}
Complexity: {business.estimated_complexity}
Agent: {agent.agent_name}
Capabilities: {', '.join(agent.capabilities)}
Provide:
- Implementation phases (at least 3-4 phases with specific tasks)
- Overall timeline estimate
- Team requirements (specific roles)
- Cost estimate range
- Potential risks (at least 3-5)
- Success metrics (KPIs)"""
if "gemini" in self.model.lower():
return await self._call_gemini_structured(system_prompt, user_prompt, api_key, ImplementationPlan)
else:
return await self._call_openai_structured(system_prompt, user_prompt, ImplementationPlan)
async def _create_summary(self, business: BusinessExtraction, agent: AgentSpecification, implementation: ImplementationPlan, api_key: str) -> str:
"""Create executive summary"""
await self._emit_log("π Generating executive summary...", "system")
system_prompt = "You are an executive summary writer. Create a concise, compelling summary."
user_prompt = f"""Create an executive summary (2-3 paragraphs) for:
Business: {business.business_name}
Industry: {business.industry}
Solution: {business.solution}
Agent: {agent.agent_name}
Timeline: {implementation.estimated_timeline}
Cost: {implementation.estimated_cost}
Make it compelling and actionable."""
if "gemini" in self.model.lower():
import google.generativeai as genai
genai.configure(api_key=api_key)
model = genai.GenerativeModel(self.model)
response = await model.generate_content_async(f"{system_prompt}\n\n{user_prompt}")
return response.text.strip()
else:
response = await self.client.chat.completions.create(
model=self.model,
messages=[
{"role": "system", "content": system_prompt},
{"role": "user", "content": user_prompt}
],
temperature=0.7,
max_tokens=500
)
return response.choices[0].message.content.strip()
async def run(self, query: str, model: str, api_key: str) -> Dict[str, Any]:
"""
Main orchestration method - runs all extraction agents in sequence.
Args:
query: User's business idea query
model: Model to use (gpt-4o, gemini-2.0-flash-exp, grok-beta, etc.)
api_key: API key for the model
Returns:
Dict with extraction results or error information
"""
try:
await self._emit_log(f"π Starting extraction with {model}...", "system")
# Setup client
self._setup_client(model, api_key)
# Step 1: Extract business information
await self._emit_log("Step 1/4: Business Analysis", "system")
business = await self._extract_business_info(query, api_key)
await self._emit_log(f"β
Identified: {business.business_name}", "success")
# Step 2: Design agent specifications
await self._emit_log("Step 2/4: Agent Design", "system")
agent = await self._extract_agent_specs(query, business, api_key)
await self._emit_log(f"β
Agent designed: {agent.agent_name}", "success")
# Step 3: Create implementation plan
await self._emit_log("Step 3/4: Implementation Planning", "system")
implementation = await self._create_implementation_plan(query, business, agent, api_key)
await self._emit_log(f"β
Timeline: {implementation.estimated_timeline}", "success")
# Step 4: Generate summary
await self._emit_log("Step 4/4: Summary Generation", "system")
summary = await self._create_summary(business, agent, implementation, api_key)
await self._emit_log("β
Extraction complete!", "success")
# Compile results
complete_extraction = CompleteExtraction(
business=business,
agent=agent,
implementation=implementation,
summary=summary
)
return {
"status": "success",
"stage": "complete",
"extraction": complete_extraction.model_dump(),
"metadata": {
"model_used": model,
"timestamp": datetime.now().isoformat(),
"session_id": self.session_id
}
}
except Exception as e:
error_msg = str(e)
await self._emit_log(f"β Error: {error_msg}", "error")
return {
"status": "error",
"stage": "extraction",
"message": error_msg,
"metadata": {
"model_used": model,
"timestamp": datetime.now().isoformat(),
"session_id": self.session_id
}
}
# ------------------------
# Testing
# ------------------------
if __name__ == "__main__":
import os
from dotenv import load_dotenv
load_dotenv()
async def test_orchestrator():
# Mock session ID
test_session = "test-session-123"
# Test query
query = "Create an AI agent for pharmacy inventory management that tracks medication expiry dates and auto-orders stock"
# Get API key from env
api_key = os.getenv("OPENAI_API_KEY", "")
if not api_key:
print("β OPENAI_API_KEY not found in .env")
return
# Run orchestrator
orchestrator = AgentOrchestrator(session_id=test_session)
result = await orchestrator.run(query, "gpt-4o", api_key)
print("\n" + "="*50)
print("EXTRACTION RESULT")
print("="*50)
print(json.dumps(result, indent=2))
asyncio.run(test_orchestrator()) |