[Bug]: No trances in agno streaming

[Anandesh-Sharma]

Contact Details

anandeshsharma@gmail.com

📦 Package Version

0.4.19

🎞️ Framework Version

agno 1.77

🔎 Describe the Bug

import os
from textwrap import dedent

from agno.agent import Agent
from agno.models.openai import OpenAIChat
from agno.tools.duckduckgo import DuckDuckGoTools
from agno.tools.newspaper4k import Newspaper4kTools

research_agent = Agent(
            model=OpenAIChat(id="o3-mini", api_key=os.getenv("OPENAI_API_KEY")),
            tools=[DuckDuckGoTools(), Newspaper4kTools()],
            description=dedent("""\
            You are an elite investigative journalist with decades of experience at the New York Times.
            Your expertise encompasses: 📰

            - Deep investigative research and analysis
            - Meticulous fact-checking and source verification
            - Compelling narrative construction
            - Data-driven reporting and visualization
            - Expert interview synthesis
            - Trend analysis and future predictions
            - Complex topic simplification
            - Ethical journalism practices
            - Balanced perspective presentation
            - Global context integration\
        """),
            instructions=dedent("""\
            1. Research Phase 🔍
            - Search for 10+ authoritative sources on the topic
            - Prioritize recent publications and expert opinions
            - Identify key stakeholders and perspectives

            2. Analysis Phase 📊
            - Extract and verify critical information
            - Cross-reference facts across multiple sources
            - Identify emerging patterns and trends
            - Evaluate conflicting viewpoints

            3. Writing Phase ✍️
            - Craft an attention-grabbing headline
            - Structure content in NYT style
            - Include relevant quotes and statistics
            - Maintain objectivity and balance
            - Explain complex concepts clearly

            4. Quality Control ✓
            - Verify all facts and attributions
            - Ensure narrative flow and readability
            - Add context where necessary
            - Include future implications
        """),
            expected_output=dedent("""\
            # {Compelling Headline} 📰

            ## Executive Summary
            {Concise overview of key findings and significance}

            ## Background & Context
            {Historical context and importance}
            {Current landscape overview}

            ## Key Findings
            {Main discoveries and analysis}
            {Expert insights and quotes}
            {Statistical evidence}

            ## Impact Analysis
            {Current implications}
            {Stakeholder perspectives}
            {Industry/societal effects}

            ## Future Outlook
            {Emerging trends}
            {Expert predictions}
            {Potential challenges and opportunities}

            ## Expert Insights
            {Notable quotes and analysis from industry leaders}
            {Contrasting viewpoints}

            ## Sources & Methodology
            {List of primary sources with key contributions}
            {Research methodology overview}

            ---
            Research conducted by AI Investigative Journalist
            New York Times Style Report
            Published: {current_date}
            Last Updated: {current_time}\
        """),
            markdown=True,
            show_tool_calls=True,
            stream=True,
            add_datetime_to_instructions=True,
        )

 agentops.init(trace_name=f"{self.name}_{self.version}")
        async def generate_model_response():
            if research_agent is None:
                # Fallback for build validation
                yield f"data: {json.dumps({'message': 'Build validation mode - agent not fully loaded'})}\n\n"
                return
                
            agent_response = await research_agent.arun(agent_input.message)
            async for chunk in agent_response:
                yield f"data: {json.dumps(chunk.content)}\n\n"

        return StreamingResponse(
          generate_model_response(),
          media_type="text/event-stream",
        )

while streaming I didn't get any traces, is there any solution to it?

🤝 Contribution

• Yes, I'd be happy to submit a pull request with these changes.
• I need some guidance on how to contribute.
• I'd prefer the AgentOps team to handle this update.