Agno Integration (mirrors examples/adapters/agno.py)
"""
Framework Adapters - Agno Integration Example
This example demonstrates how to integrate Syntha with the Agno framework
for flexible multi-agent coordination with shared context.
Prerequisites:
- pip install agno (if using real Agno framework)
- Set OPENAI_API_KEY environment variable
Copy and run this code to see Agno integration in action!
"""
import os
from syntha import ContextMesh, ToolHandler, build_system_prompt, SynthaFrameworkError
def simulate_agno_agent(name, tools, prompt):
"""
Simulate Agno agent for demonstration.
Replace this with actual Agno agent in real usage.
"""
print(f"🤖 [SIMULATED] Agno Agent '{name}' Execution")
# Extract tool names robustly for both Agno Function objects and schema dicts
tool_names = []
for t in tools[:3]:
if isinstance(t, dict) and "name" in t:
tool_names.append(t["name"])
elif isinstance(t, dict) and "function" in t and isinstance(t["function"], dict):
tool_names.append(t["function"].get("name", "unknown"))
else:
name_attr = getattr(t, "name", None)
tool_names.append(name_attr or getattr(t, "__name__", "unknown"))
print(f" Available tools: {tool_names}...")
print(f" Prompt length: {len(prompt)} characters")
# Simulate agent using tools
return {
"response": f"Agent {name} analyzed the context and completed the task.",
"tools_used": ["get_context", "push_context"],
"context_shared": True,
}
def real_agno_example():
"""
Example of real Agno integration (commented out for demo).
Uncomment and modify for actual usage.
"""
# Uncomment these lines for real Agno usage:
# from agno import Agent
#
# agent = Agent(
# name="SynthaAgent",
# tools=agno_tools,
# instructions=system_prompt,
# model="gpt-4"
# )
#
# result = agent.run(user_message)
# return result
def main():
print("🚀 Framework Adapters - Agno Integration")
print("=" * 45)
# Check for API key (for real usage)
api_key = os.getenv("OPENAI_API_KEY")
if api_key:
print("✅ OpenAI API key found for Agno")
else:
print("⚠️ No OpenAI API key found - using simulation mode")
print(" Set OPENAI_API_KEY environment variable for real usage")
# 1. Set up Syntha
context = ContextMesh(user_id="content_team")
# Create handlers for different content agents
writer_handler = ToolHandler(context, "ContentWriter")
editor_handler = ToolHandler(context, "ContentEditor")
seo_handler = ToolHandler(context, "SEOSpecialist")
# Add content project context
context.push(
"content_strategy",
{
"campaign": "Product Launch Blog Series",
"target_audience": "B2B software buyers",
"content_types": ["blog_posts", "case_studies", "whitepapers"],
"publishing_schedule": "3 posts per week",
"duration": "8 weeks",
},
)
context.push(
"brand_guidelines",
{
"tone": "professional but approachable",
"voice": "expert and helpful",
"keywords": ["automation", "efficiency", "productivity", "ROI"],
"avoid_terms": ["cheap", "basic", "simple"],
},
)
context.push(
"seo_targets",
{
"primary_keywords": ["business automation", "workflow optimization"],
"secondary_keywords": ["productivity tools", "process improvement"],
"target_length": "1500-2000 words",
"internal_links": 3,
"external_links": 2,
},
)
print("✅ Content strategy context added to mesh")
# 2. Get Agno-compatible tools for each agent (fallback to schemas if Agno not installed)
try:
writer_tools = writer_handler.get_tools_for_framework("agno")
editor_tools = editor_handler.get_tools_for_framework("agno")
seo_tools = seo_handler.get_tools_for_framework("agno")
print(f"🔧 Created Agno tools for 3 agents:")
print(f" Writer: {len(writer_tools)} tools")
print(f" Editor: {len(editor_tools)} tools")
print(f" SEO: {len(seo_tools)} tools")
except SynthaFrameworkError as e:
print("⚠️ Agno not available (pip install agno). Falling back to raw schemas for demo.")
writer_tools = writer_handler.get_schemas()
editor_tools = editor_handler.get_schemas()
seo_tools = seo_handler.get_schemas()
print(f"🔧 Using schemas instead:")
print(f" Writer: {len(writer_tools)} schemas")
print(f" Editor: {len(editor_tools)} schemas")
print(f" SEO: {len(seo_tools)} schemas")
# 3. Build context-aware prompts for each agent
writer_prompt = build_system_prompt("ContentWriter", context)
editor_prompt = build_system_prompt("ContentEditor", context)
seo_prompt = build_system_prompt("SEOSpecialist", context)
# 4. Simulate Agno agents working together
print("\n🔄 Multi-agent content workflow:")
# Writer creates content
writer_result = simulate_agno_agent("ContentWriter", writer_tools, writer_prompt)
print(f" ✍️ Writer: {writer_result['response']}")
# Writer shares draft with team
writer_handler.handle_tool_call(
"push_context",
key="draft_article",
value={
"title": "10 Ways Business Automation Transforms Your Workflow",
"word_count": 1750,
"status": "first_draft",
"sections": [
"Introduction",
"Benefits",
"Implementation",
"Case Studies",
"Conclusion",
],
},
topics=["content", "review"],
)
# Editor reviews and provides feedback
editor_result = simulate_agno_agent("ContentEditor", editor_tools, editor_prompt)
print(f" ✏️ Editor: {editor_result['response']}")
# Editor shares feedback
editor_handler.handle_tool_call(
"push_context",
key="editorial_feedback",
value={
"overall_rating": "good",
"suggested_changes": [
"Strengthen introduction hook",
"Add more specific examples in section 2",
"Improve conclusion call-to-action",
],
"grammar_issues": 3,
"structure_score": 8,
},
topics=["content", "feedback"],
)
# SEO Specialist optimizes content
seo_result = simulate_agno_agent("SEOSpecialist", seo_tools, seo_prompt)
print(f" 🎯 SEO: {seo_result['response']}")
# SEO shares optimization recommendations
seo_handler.handle_tool_call(
"push_context",
key="seo_analysis",
value={
"keyword_density": "optimal",
"meta_description": "Business automation tools can transform your workflow efficiency by 40%. Learn 10 proven strategies to optimize processes and boost ROI.",
"suggested_tags": ["automation", "productivity", "business-tools"],
"readability_score": 85,
"seo_score": 92,
},
topics=["content", "seo", "final"],
)
# 5. Show how all agents can access shared context
print("\n📊 Final content status:")
for agent_name, handler in [
("Writer", writer_handler),
("Editor", editor_handler),
("SEO", seo_handler),
]:
agent_context = handler.handle_tool_call("get_context")
context_keys = list(agent_context["context"].keys())
print(f" {agent_name} sees: {len(context_keys)} context items")
# 6. Demonstrate Agno-specific features
print("\n🎯 Agno-specific capabilities:")
print(" - Tools formatted for Agno's agent system")
print(" - Context-aware prompts for each agent role")
print(" - Seamless multi-agent coordination")
print(" - Flexible tool access control")
# 7. Show workflow completion
final_context = context.get_all_for_agent("ContentWriter")
workflow_items = [
key
for key in final_context.keys()
if key in ["draft_article", "editorial_feedback", "seo_analysis"]
]
print(f"\n✅ Content workflow completed:")
print(f" Workflow items: {workflow_items}")
print(" Ready for publication with full team input")
print("\n💡 Next steps for real integration:")
print("1. Install: pip install agno")
print("2. Set environment variable: export OPENAI_API_KEY='your-key-here'")
print("3. Replace simulate_agno_agent() with real Agno agents")
print("4. Use get_tools_for_framework('agno') for tool integration")
print("5. Build complex multi-agent workflows with Agno's flexibility")
print("\n✅ Agno integration example complete!")
if __name__ == "__main__":
main()
See also: - Multi-Agent Setup - Framework Adapters