Autonomous-Research-Agent/app.py at main · esmailnaim/Autonomous-Research-Agent · GitHub

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
import streamlit as st
from agent import create_research_agent, run_research
import config

# Page configuration
st.set_page_config(
    page_title="DeepSeek Research Agent",
    page_icon="🔍",
    layout="wide",
)

# Initialize session state
if "agent" not in st.session_state:
    st.session_state.agent = None
if "research_history" not in st.session_state:
    st.session_state.research_history = []


def main():
    """Main Streamlit application."""
    st.title("🔍 DeepSeek Research Agent")
    st.markdown("An autonomous research agent powered by DeepSeek and Tavily search.")

    # Sidebar configuration
    with st.sidebar:
        st.header("Configuration")
        st.info(f"Model: {config.DEEPSEEK_MODEL}")
        st.info(f"Max Results: {config.TAVILY_MAX_RESULTS}")
        st.info(f"Search Depth: {config.TAVILY_SEARCH_DEPTH}")

        if st.button("Initialize Agent"):
            with st.spinner("Initializing agent..."):
                try:
                    st.session_state.agent = create_research_agent()
                    st.success("Agent initialized successfully!")
                except Exception as e:
                    st.error(f"Error initializing agent: {str(e)}")

    # Main content area
    st.header("Research Query")

    # Query input
    query = st.text_area(
        "Enter your research query:",
        height=100,
        placeholder="e.g., What are the latest developments in AI reasoning?",
    )

    col1, col2 = st.columns([1, 4])
    with col1:
        research_button = st.button("🔍 Research", type="primary", use_container_width=True)
    with col2:
        clear_button = st.button("Clear History", use_container_width=True)

    if clear_button:
        st.session_state.research_history = []
        st.rerun()

    # Run research
    if research_button and query:
        if st.session_state.agent is None:
            st.warning("Please initialize the agent first from the sidebar.")
        else:
            with st.spinner("Researching..."):
                try:
                    result = run_research(query, st.session_state.agent)

                    # Store in history
                    st.session_state.research_history.append({
                        "query": query,
                        "result": result,
                    })

                    # Display results
                    st.success("Research completed!")

                    # Show report draft (main output)
                    if result.get("report_draft"):
                        st.subheader("📄 Research Report")
                        st.markdown(result["report_draft"])

                    # Show research progress
                    col1, col2, col3 = st.columns(3)
                    with col1:
                        st.metric("Sections", len(result.get("sections", [])))
                    with col2:
                        st.metric("Search Queries", len(result.get("search_queries", [])))
                    with col3:
                        confidence = result.get("confidence_score", 0)
                        if confidence:
                            st.metric("Confidence", f"{confidence*100:.1f}%")

                    # Show sections and research results
                    if result.get("sections"):
                        st.subheader("📚 Research Sections")
                        for section in result.get("sections", []):
                            with st.expander(f"Section: {section}"):
                                if section in result.get("research_results", {}):
                                    for res in result["research_results"][section]:
                                        st.markdown(f"**Query:** {res.get('query', 'N/A')}")
                                        st.markdown(f"**Summary:** {res.get('summary', 'N/A')}")
                                else:
                                    st.info("No research data for this section yet.")

                    # Show reflection feedback
                    if result.get("reflection_feedback"):
                        st.subheader("🤔 Reflection & Quality Assessment")
                        st.info(result["reflection_feedback"])

                    # Show sources
                    if result.get("sources"):
                        st.subheader("🔗 Sources")
                        for i, source in enumerate(result["sources"][:10], 1):
                            st.markdown(f"{i}. [{source.get('title', 'Untitled')}]({source.get('url', '#')})")

                    # Show raw messages (for debugging)
                    with st.expander("🔍 Debug: Messages"):
                        for message in result.get("messages", []):
                            if hasattr(message, "content"):
                                st.markdown(f"**{message.__class__.__name__}:**")
                                st.text(message.content[:500])  # Truncate for display

                except Exception as e:
                    st.error(f"Error during research: {str(e)}")
                    st.exception(e)

    # Research history
    if st.session_state.research_history:
        st.header("Research History")
        for i, item in enumerate(reversed(st.session_state.research_history), 1):
            with st.expander(f"Query {i}: {item['query'][:50]}..."):
                st.markdown(f"**Query:** {item['query']}")
                st.json(item["result"])


if __name__ == "__main__":
    main()