allankltsn
Community@allankltsn
Comprehensive Agno agent engineering curriculum covering RAG, guardrails, memory, multi-agent teams, testing, and serverless AWS deployment with FastAPI and SAM.
Agent Skills by allankltsn
Showing 26 vetted skills indexed across 1 GitHub repositories.
allankltsnagno-knowledge-rag
Implement retrieval-augmented generation in Agno agents with pluggable vector databases.
allankltsnagno-guardrails
Configures mandatory input guardrails for Agno agents including PII, prompt injection, moderation, and scope checks.
allankltsnagno-security
Applies OWASP security practices to Agno agents and FastAPI production APIs.
allankltsnagno-aws
Defines AWS SAM infrastructure patterns and boto3 best practices for deploying Agno agents.
allankltsnnuageit-slide-deck
Generates interactive single-file HTML slide decks with a reusable design system and navigation engine.
allankltsnagno-reasoning
Implements explicit reasoning steps in Agno agents using reasoning models, chain-of-thought, and ReasoningTools.
allankltsnagno-project-structure
Defines the standard folder layout and file structure for Agno agent projects.
allankltsnagno-development-master
Guides creation of isolated Agno v2 study agents across progressive learning levels.
allankltsnagno-prompt-system
Writes and adapts Agno agent system prompts per model family with S3-backed prompt storage.
allankltsnagno-secrets
Resolves secrets for Agno agents via environment variables with AWS Secrets Manager fallback.
allankltsnagno-memory-sessions
Configure Agno agent memory and session persistence with DynamoDB or SQLite backends.
allankltsnagno-project-architecture-solid
Defines layered folder structure and SOLID principles for Agno agent projects deployed serverless on AWS.
allankltsnagno-providers
Configure Agno LLM providers with environment-driven model IDs and parameters.
allankltsnagno-testing-strategy
Test and optimize Agno agents with deterministic fakes, evals, and parameter search.
allankltsnagno-teams
Create multi-agent teams in Agno with a leader model that delegates tasks dynamically.
allankltsnagno-lambda-apigateway
Deploy Agno agents as FastAPI REST APIs on AWS Lambda behind API Gateway.
allankltsnagno-code-quality-tools
Configures deterministic local code quality tooling for Python Agno agents using Ruff, mypy, and pre-commit.
allankltsnagno-agentos-eval-deploy
Serve Agno agents as FastAPI services with tracing, evaluation, and AWS EC2 deployment.
allankltsnagno-local-dev-environment
Configure a docker-compose local environment simulating AWS, LLM providers, and observability for Agno agents.
allankltsnagno-multimodal
Build Agno agents that process image and audio inputs alongside text prompts.
allankltsnagno-observability
Instrument Agno v2 agents with OpenTelemetry tracing, structured logging, and metrics.
allankltsnagno-workflows-basic
Orchestrates sequential Agno Workflows with Steps executed by Agents, Teams, or functions.
allankltsnagno-workflows-advanced
Implements Agno workflow control flow with Parallel, Condition, Loop, and Router steps.
allankltsnagno-tools
Implement ready-made and custom Python tools for Agno agents using the @tool decorator.
Frequently Asked Questions About allankltsn
FAQPage SchemaWhat tasks can I accomplish with allankltsn's Agno skills?▼
You can build Agno agents end-to-end: system prompts per provider, RAG with pluggable vector databases, memory and sessions, multi-agent teams, basic and advanced orchestration, multimodal I/O, custom tools, guardrails, observability, automated testing and optimization, and serverless AWS deployment.
Who are these Agno skills designed for?▼
They target engineers studying or shipping Agno agents to production: backend developers exposing agents via FastAPI on Lambda, ML engineers tuning RAG and sampling parameters with Optuna and DSPy, and security reviewers applying OWASP LLM and API Top 10 controls.
How do I set up a local development environment for these agents?▼
Use the environment-setup skill with uv for isolated per-agent projects and env vars, then the local-dev-environment skill's docker-compose stack: LocalStack for AWS, Ollama as local model provider, SigNoz or Grafana/Tempo/Loki/Prometheus for OTLP observability, Langfuse for tracing, and AgentOS for monitoring.
How are secrets and configuration handled in these agents?▼
The agno-secrets skill defines a fallback resolution pipe: standard environment variable first, then AWS Secrets Manager, failing fast with an exception when a required variable is missing from every source. Model selection and provider parameters are also env-driven with predefined defaults.
What testing and quality practices do these skills enforce?▼
Three test layers (deterministic transport, agent behavior, real system), golden-set evals with deterministic and LLM-as-judge scorers, Optuna/DSPy/promptfoo/Ragas optimization, plus local quality gates: Ruff, mypy, Bandit, pip-audit, gitleaks, import-linter, Radon, pytest-cov, Hypothesis, mutmut, and pre-commit.