multi agent
Browse files
app.py
CHANGED
|
@@ -144,8 +144,8 @@ With the right competitive research, you don’t just react to the market—you
|
|
| 144 |
|
| 145 |
gr.Examples(
|
| 146 |
[
|
| 147 |
-
|
| 148 |
-
"Docker Containers", "REST API", "Python"
|
| 149 |
],
|
| 150 |
[in_verbatim]
|
| 151 |
)
|
|
|
|
| 144 |
|
| 145 |
gr.Examples(
|
| 146 |
[
|
| 147 |
+
|
| 148 |
+
"Docker Containers", "REST API", "Python"
|
| 149 |
],
|
| 150 |
[in_verbatim]
|
| 151 |
)
|
multi.py
CHANGED
|
@@ -5,20 +5,22 @@ from agno.tools.duckduckgo import DuckDuckGoTools
|
|
| 5 |
from agno.models.ollama import Ollama
|
| 6 |
from agno.models.groq import Groq
|
| 7 |
import os
|
| 8 |
-
chat=Groq(id='llama-3.
|
|
|
|
| 9 |
# Create individual specialized agents
|
| 10 |
researcher = Agent(
|
| 11 |
name="Researcher",
|
| 12 |
-
role="Expert at finding information by breaking the structure into components",
|
| 13 |
-
tools=[DuckDuckGoTools()],
|
| 14 |
show_tool_calls=True,
|
|
|
|
| 15 |
model=chat, #OpenAIChat("gpt-4o"),
|
| 16 |
#debug_mode=True,
|
| 17 |
)
|
| 18 |
|
| 19 |
engineer = Agent(
|
| 20 |
name="Security Engineer",
|
| 21 |
-
role="Security Expert at writing clear, engaging content for hands-on best practices, and common pitfalls with solution",
|
| 22 |
model=chat, #OpenAIChat("gpt-4o"),
|
| 23 |
)
|
| 24 |
|
|
@@ -33,14 +35,14 @@ content_team = Team(
|
|
| 33 |
)
|
| 34 |
def bestPractice(topic):
|
| 35 |
r = content_team.run(topic)
|
| 36 |
-
return r.messages[-1].content
|
| 37 |
|
| 38 |
if __name__=='__main__':
|
| 39 |
from pprint import pprint
|
| 40 |
from agno.utils.pprint import pprint_run_response
|
| 41 |
r=content_team.run("Docker Containers")
|
| 42 |
pprint_run_response(r, markdown=True)
|
| 43 |
-
print([m
|
| 44 |
|
| 45 |
|
| 46 |
print("")
|
|
|
|
| 5 |
from agno.models.ollama import Ollama
|
| 6 |
from agno.models.groq import Groq
|
| 7 |
import os
|
| 8 |
+
#chat=Groq(id='llama-3.3-70b-versatile') if os.getenv("GROQ_API_KEY") else
|
| 9 |
+
chat=Ollama(id="qwen2.5")
|
| 10 |
# Create individual specialized agents
|
| 11 |
researcher = Agent(
|
| 12 |
name="Researcher",
|
| 13 |
+
role="Expert at finding information by breaking the structure into components ie) architecture, code, algorithm, linux system",
|
| 14 |
+
tools=[DuckDuckGoTools(fixed_max_results=3)],
|
| 15 |
show_tool_calls=True,
|
| 16 |
+
tool_call_limit=1,
|
| 17 |
model=chat, #OpenAIChat("gpt-4o"),
|
| 18 |
#debug_mode=True,
|
| 19 |
)
|
| 20 |
|
| 21 |
engineer = Agent(
|
| 22 |
name="Security Engineer",
|
| 23 |
+
role="Security Expert at writing short, clear, engaging content for hands-on best practices, and common pitfalls with solution",
|
| 24 |
model=chat, #OpenAIChat("gpt-4o"),
|
| 25 |
)
|
| 26 |
|
|
|
|
| 35 |
)
|
| 36 |
def bestPractice(topic):
|
| 37 |
r = content_team.run(topic)
|
| 38 |
+
return [m for m in r.messages if m.role == 'assistant'][-1].content
|
| 39 |
|
| 40 |
if __name__=='__main__':
|
| 41 |
from pprint import pprint
|
| 42 |
from agno.utils.pprint import pprint_run_response
|
| 43 |
r=content_team.run("Docker Containers")
|
| 44 |
pprint_run_response(r, markdown=True)
|
| 45 |
+
print([m for m in r.messages if m.role == 'assistant'][-1].content)
|
| 46 |
|
| 47 |
|
| 48 |
print("")
|