skills.rest
allankltsn
allankltsnCOMMUNITY
Explore all skills by this author →
GitHub

agno-agentos-eval-deploy

Serve Agno agents as FastAPI services with tracing, evaluation, and AWS EC2 deployment.

Updated Sep 10, 2026
Search Tags:#deployment#fastapi#tracing#agentos#agno#agent evaluation#aws ec2
One-click install
npx skills add https://github.com/allankltsn/setup_kiro --skill agno-agentos-eval-deploy-allankltsn
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: agno-agentos-eval-deploy
Source: https://github.com/allankltsn/setup_kiro/tree/main/.kiro/skills/agno-agentos-eval-deploy
Command: npx skills add https://github.com/allankltsn/setup_kiro --skill agno-agentos-eval-deploy-allankltsn

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires agno, uvicorn.

What problem does it solve? Turning a local Agno agent into a production service requires wiring up an API, a UI for testing, tracing for debugging, and a deployment path, which is error-prone when done ad hoc. ## Core Features & Use Cases - AgentOS API packaging: Wrap agents, teams, and workflows in a FastAPI app with AgentOS, served via uvicorn on port 7777. - Web UI and tracing: Connect os.agno.com to the local endpoint to chat, inspect sessions, and review step-by-step traces of tool calls and reasoning. - Evaluation workflow: Compare agent versions using the same inputs and review persisted sessions to measure quality and regressions. - AWS EC2 deployment: Follow a documented flow covering security groups, environment variables, systemd, Nginx with HTTPS, and pinned versions. - Use Case: You built a chat agent with Agno and want to expose it as an API, debug its traces in the Agno UI, then deploy it to a small EC2 instance behind HTTPS. ## Quick Start Serve my Agno agent with AgentOS, enable tracing, and walk me through deploying it to an EC2 instance.

Frequently Asked Questions about agno-agentos-eval-deploy

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I serve an Agno agent as a FastAPI API?▼

Create an AgentOS instance with your agents, teams, or workflows, call get_app() to obtain a FastAPI app, and run agent_os.serve(app="main:app", reload=True). Install the dependencies with uv add "agno[fastapi]" uvicorn.

How do I trace and debug Agno agent executions?▼

Enable tracing=True on AgentOS, run the server on localhost:7777, then connect os.agno.com to the local endpoint. The UI exposes Chat, Sessions, and Traces views to inspect tool calls, reasoning, and outputs step by step.

Can AgentOS host multiple agents, teams, and workflows together?▼

Yes, AgentOS accepts agents, teams, and workflows lists in the same instance and automatically discovers the components they use, such as models, tools, and databases. Use a separate database per environment to isolate sessions and memory.

How do I deploy an Agno AgentOS app to AWS EC2?▼

Launch an EC2 instance, open port 7777 in the security group, install Python 3.9+ and uv, clone the project, and run uv sync. Run AgentOS under systemd behind an Nginx HTTPS proxy and store API keys as environment variables.

What security practices apply when exposing AgentOS publicly?▼

Never hardcode API keys; use environment variables or a secrets manager. Restrict the security group, serve HTTPS through a reverse proxy instead of exposing AgentOS directly, pin versions with uv.lock, and consider JWT-based RBAC before multiuser access.

Related Skills

Explore More Software Engineering Skills→
aeonbridgeaeonbridge

agno

Build, deploy, and manage multi-agent AI systems with Agno and AgentOS workflows.

Official
Advanced
juanlamadrid20juanlamadrid20

agent-mlops

Automate production deployment and monitoring of AI agents on Databricks.

Community
Advanced
towardsthecloudtowardsthecloud
1.8k

systematic-debugging

Identify root causes of bugs through a four-phase investigation process.

Official
Advanced
infantesromeroadrianinfantesromeroadrian

production

Deploy ML models to production with FastAPI, Docker, and Kubernetes templates.

Community
Advanced