File size: 7,513 Bytes
fb09fd5
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
import gradio as gr
import os
from smolagents import CodeAgent, ToolCollection, Tool
from smolagents.models import HfApiModel, LiteLLMModel

# Default model to use
DEFAULT_MODEL = "Qwen/Qwen2.5-Coder-7B-Instruct"

# Tool descriptions for the UI
TOOL_DESCRIPTIONS = {
    "Hub Collections": "Add tool collections from Hugging Face Hub.",
    "Spaces": "Add tools from Hugging Face Spaces.",
}


# Function to create an agent with selected tools
def create_agent(model_name, hub_tool=None, space_tool=None):
    tools = []

    # Add tool from Hub if provided
    if hub_tool:
        try:
            hub_collection = ToolCollection.from_hub(collection_slug=hub_tool)
            tools.extend(hub_collection.tools)
        except Exception as e:
            print(f"Error loading Hub tool: {e}")

    # Add tool from Space if provided
    if space_tool:
        try:
            space_tool_obj = Tool.from_space(
                space_id=space_tool,
                name=f"space_{space_tool.replace('/', '_')}",
                description=f"Tool from Hugging Face Space: {space_tool}",
            )
            tools.append(space_tool_obj)
        except Exception as e:
            print(f"Error loading Space tool: {e}")

    # Create and return the agent
    try:
        # Try to use HfApiModel first
        model = HfApiModel(model_id=model_name)
        return CodeAgent(tools=tools, model=model)
    except Exception:
        # Fall back to LiteLLMModel if HfApiModel fails
        try:
            model = LiteLLMModel(model_id=model_name)
            return CodeAgent(tools=tools, model=model)
        except Exception as e:
            print(f"Error creating agent: {e}")
            return None


# Main application
def main():
    with gr.Blocks(title="AI Agent Builder") as app:
        gr.Markdown("# AI Agent Builder with SmolaGents")
        gr.Markdown(
            "Build your own AI agent by selecting tools from Hugging Face Hub and Spaces."
        )

        with gr.Tabs():
            with gr.TabItem("Build Agent"):
                with gr.Row():
                    with gr.Column(scale=1):
                        # Model selection
                        model_input = gr.Textbox(
                            label="Model Name",
                            placeholder="Enter model name or ID",
                            value=DEFAULT_MODEL,
                        )

                        # Hub tool input
                        hub_tool_input = gr.Textbox(
                            label="Add Tool Collection from Hub (collection slug)",
                            placeholder="e.g., huggingface-tools/diffusion-tools-...",
                        )

                        # Space tool input
                        space_tool_input = gr.Textbox(
                            label="Add Tool from Space (space ID)",
                            placeholder="e.g., black-forest-labs/FLUX.1-schnell",
                        )

                        # Create agent button
                        create_button = gr.Button("Create Agent")

                        # Status message
                        status_msg = gr.Markdown("")

                    with gr.Column(scale=2):
                        # Chat interface for the agent
                        chatbot = gr.Chatbot(label="Agent Chat")
                        msg = gr.Textbox(label="Your message")

                        with gr.Row():
                            clear = gr.Button("Clear Chat")
                            regenerate = gr.Button("Regenerate Response")

            with gr.TabItem("Tool Descriptions"):
                tool_descriptions_md = """
## Hugging Face Hub Tool Collections

You can add tool collections from Hugging Face Hub by providing the collection slug.
Example: `huggingface-tools/diffusion-tools-6630bb19a942c2306a2cdb6f`

## Hugging Face Spaces as Tools

You can add tools from Hugging Face Spaces by providing the space ID.
Example: `black-forest-labs/FLUX.1-schnell`

This allows you to use any Gradio app on Hugging Face Spaces as a tool for your agent.
                """

                gr.Markdown(tool_descriptions_md)

        # Agent state
        agent_state = gr.State(None)
        last_message = gr.State("")

        # Event handlers
        def on_create_agent(model, hub_tool, space_tool):
            if not model:
                return None, [], "", "⚠️ Please enter a model name."

            if not hub_tool and not space_tool:
                return None, [], "", "⚠️ Please add at least one tool from Hub or Space."

            agent = create_agent(model, hub_tool, space_tool)

            if agent is None:
                return (
                    None,
                    [],
                    "",
                    "❌ Failed to create agent. Check console for details.",
                )

            tools_info = []
            if hub_tool:
                tools_info.append(f"Hub collection: {hub_tool}")
            if space_tool:
                tools_info.append(f"Space: {space_tool}")

            tools_str = " | ".join(tools_info)

            return (
                agent,
                [],
                "",
                f"✅ Agent created successfully with {model}! ({tools_str})",
            )

        create_button.click(
            on_create_agent,
            inputs=[model_input, hub_tool_input, space_tool_input],
            outputs=[agent_state, chatbot, msg, status_msg],
        )

        def on_message(message, chat_history, agent, last_msg):
            if not message:
                return "", chat_history, last_msg

            if agent is None:
                chat_history.append((message, "Please create an agent first."))
                return "", chat_history, last_msg

            try:
                response = agent.run(message, reset=False)
                chat_history.append((message, response))
                return "", chat_history, message
            except Exception as e:
                error_msg = f"Error: {str(e)}"
                chat_history.append((message, error_msg))
                return "", chat_history, message

        msg.submit(
            on_message,
            inputs=[msg, chatbot, agent_state, last_message],
            outputs=[msg, chatbot, last_message],
        )

        def on_regenerate(chat_history, agent, last_msg):
            if not chat_history or not last_msg or agent is None:
                return chat_history, last_msg

            try:
                # Remove the last exchange
                if chat_history:
                    chat_history.pop()

                # Regenerate the response
                response = agent.run(last_msg, reset=False)
                chat_history.append((last_msg, response))
                return chat_history, last_msg
            except Exception as e:
                error_msg = f"Error regenerating response: {str(e)}"
                chat_history.append((last_msg, error_msg))
                return chat_history, last_msg

        regenerate.click(
            on_regenerate,
            inputs=[chatbot, agent_state, last_message],
            outputs=[chatbot, last_message],
        )

        def on_clear():
            return None, [], "", "Agent cleared. Create a new one to continue."

        clear.click(on_clear, outputs=[agent_state, chatbot, last_message, status_msg])

    return app


if __name__ == "__main__":
    app = main()
    app.launch()