What problem does it solve? Developing an Agno agent against real AWS services, paid LLM providers, and production observability stacks is costly and slow. This Skill sets up a fully local docker-compose environment that emulates every third-party dependency so the agent code runs unchanged between local development and production. ## Core Features & Use Cases - LocalStack emulation: Simulates API Gateway, Lambda, DynamoDB, Secrets Manager, S3, and SSM with idempotent bootstrap scripts, letting you rehearse SAM deploys without an AWS account. - Ollama local LLM fallback: Runs a local model provider for offline, zero-cost inference alongside remote providers. - Pluggable observability backends: Compose profiles switch between SigNoz (default) and Grafana+Tempo+Loki+Prometheus, plus optional Langfuse for LLM tracing, all receiving OTLP telemetry from the agent. - AgentOS monitoring UI: Serves the agent through Agno AgentOS on port 7777 to inspect runs, sessions, tokens, and traces while developing. - Use Case: While building a support agent, run docker compose up -d, point .env.local at LocalStack and the local OTLP collector, then watch each agent run in AgentOS and traces in SigNoz before deploying to AWS Lambda. ## Quick Start Ask the AI to set up the local Agno development environment with docker-compose including LocalStack, Ollama, AgentOS, and a SigNoz observability backend for my agent project.
