File size: 6,625 Bytes
cc98b5f
 
 
 
 
 
 
 
 
 
 
7a4e132
cc98b5f
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
7a4e132
 
cc98b5f
 
 
 
7a4e132
cc98b5f
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
7a4e132
cc98b5f
7a4e132
 
cc98b5f
 
 
 
 
 
 
 
7a4e132
 
cc98b5f
 
 
 
 
 
 
 
7a4e132
 
cc98b5f
 
 
 
 
 
 
 
 
7a4e132
cc98b5f
7a4e132
 
cc98b5f
 
 
 
7a4e132
 
cc98b5f
 
 
 
 
7a4e132
 
cc98b5f
 
 
 
 
 
 
 
 
7a4e132
 
 
cc98b5f
 
 
 
 
 
 
 
 
 
 
7a4e132
cc98b5f
 
 
 
 
 
7a4e132
cc98b5f
 
7a4e132
cc98b5f
7a4e132
cc98b5f
7a4e132
cc98b5f
7a4e132
cc98b5f
7a4e132
 
cc98b5f
 
7a4e132
cc98b5f
7a4e132
cc98b5f
 
 
7a4e132
cc98b5f
7a4e132
cc98b5f
 
 
7a4e132
 
 
cc98b5f
 
7a4e132
 
cc98b5f
7a4e132
cc98b5f
7a4e132
 
cc98b5f
 
 
 
 
 
 
 
 
 
 
 
7a4e132
cc98b5f
 
 
7a4e132
cc98b5f
7a4e132
 
 
 
 
 
 
 
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
import os
import textwrap

# ─── CrewAI core ─────────────────────────────────────────────────────────────
from crewai import Agent, Task, Crew, Process, LLM
from crewai_tools import SerperDevTool, ScrapeWebsiteTool

# ─── UI ──────────────────────────────────────────────────────────────────────
import gradio as gr

# =============================================================================
# API KEYS
# =============================================================================
OPENAI_API_KEY = os.environ.get("OPENAI_API_KEY", "")
SERPER_API_KEY = os.environ.get("SERPER_API_KEY", "")

if not OPENAI_API_KEY:
    raise EnvironmentError(
        "❌ OPENAI_API_KEY not found. "
        "Add it in Space β†’ Settings β†’ Variables and secrets."
    )

if not SERPER_API_KEY:
    raise EnvironmentError(
        "❌ SERPER_API_KEY not found. "
        "Add it in Space β†’ Settings β†’ Variables and secrets."
    )

os.environ["OPENAI_API_KEY"] = OPENAI_API_KEY
os.environ["SERPER_API_KEY"] = SERPER_API_KEY

# =============================================================================
# LLM
# =============================================================================
llm = LLM(
    model="gpt-4o-mini",
    temperature=0.4,
    max_tokens=2000,
)

# =============================================================================
# TOOLS
# =============================================================================
search_tool = SerperDevTool()
scrape_tool = ScrapeWebsiteTool()

# =============================================================================
# MEMORY
# =============================================================================
MEMORY_DIR = "/tmp/crewai_memory"
os.makedirs(MEMORY_DIR, exist_ok=True)

EMBEDDER_CONFIG = {
    "provider": "openai",
    "config": {
        "model": "text-embedding-3-small",
        "api_key": OPENAI_API_KEY,
    },
}

# =============================================================================
# AGENTS
# =============================================================================
researcher = Agent(
    role="Senior Research Analyst",
    goal=(
        "Find the most accurate, up-to-date, and relevant information on the given topic."
    ),
    backstory="Veteran research analyst with deep investigation skills.",
    tools=[search_tool, scrape_tool],
    llm=llm,
    verbose=True,
    allow_delegation=False,
    max_iter=5,
)

writer = Agent(
    role="Professional Content Writer",
    goal="Write a compelling structured article from research.",
    backstory="Expert in long-form and technical writing.",
    llm=llm,
    verbose=True,
    allow_delegation=False,
    max_iter=4,
)

editor = Agent(
    role="Senior Editor",
    goal="Polish article to publication quality.",
    backstory="Meticulous editor ensuring clarity and consistency.",
    llm=llm,
    verbose=True,
    allow_delegation=False,
    max_iter=3,
)

# =============================================================================
# TASK BUILDER
# =============================================================================
def build_tasks(topic: str, audience: str, tone: str, length: str):
    research_task = Task(
        description=f"Research topic: {topic}",
        expected_output="Structured research summary",
        agent=researcher,
    )

    writing_task = Task(
        description=f"Write article on: {topic}",
        expected_output="Full article",
        agent=writer,
        context=[research_task],
    )

    editing_task = Task(
        description="Edit article to final form",
        expected_output="Final polished article",
        agent=editor,
        context=[writing_task],
    )

    return [research_task, writing_task, editing_task]

# =============================================================================
# CREW RUNNER
# =============================================================================
def run_crew(topic: str, audience: str, tone: str, length: str):
    if not topic.strip():
        return "Please enter a topic."

    try:
        tasks = build_tasks(topic, audience, tone, length)

        crew = Crew(
            agents=[researcher, writer, editor],
            tasks=tasks,
            process=Process.sequential,
            verbose=True,
            memory=True,
            embedder=EMBEDDER_CONFIG,
            memory_config={"long_term": {"storage_path": MEMORY_DIR}},
        )

        result = crew.kickoff()
        return str(result)

    except Exception as e:
        return f"Error: {str(e)}"

# =============================================================================
# UI
# =============================================================================
with gr.Blocks() as demo:

    gr.Markdown("# Multi-Agent Article Generator")

    with gr.Row():

        with gr.Column():
            topic_input = gr.Textbox(label="Topic", lines=3)

            audience_input = gr.Dropdown(
                ["General Public", "Business", "Students"],
                value="General Public",
                label="Audience",
            )

            tone_input = gr.Dropdown(
                ["Informative", "Conversational", "Academic"],
                value="Informative",
                label="Tone",
            )

            length_input = gr.Radio(
                ["Short", "Medium", "Long"],
                value="Medium",
                label="Length",
            )

            submit_btn = gr.Button("Generate")
            clear_btn = gr.Button("Clear")

        with gr.Column():
            output_box = gr.Textbox(
                label="Output",
                lines=30,
                interactive=False,
            )

    submit_btn.click(
        fn=run_crew,
        inputs=[topic_input, audience_input, tone_input, length_input],
        outputs=output_box,
    )

    clear_btn.click(
        fn=lambda: ("", ""),
        inputs=[],
        outputs=[topic_input, output_box],
    )

# =============================================================================
# LAUNCH (UPDATED FOR GRADIO 6)
# =============================================================================
demo.launch(
    server_name="0.0.0.0",
    server_port=7860,
    theme=gr.themes.Soft(primary_hue="blue", secondary_hue="slate"),
    css="""
        footer { display: none !important; }
    """,
)