aboyalejandro

Run the live agent against real scenarios, capture every tool call, score with judges that see what the agent actually did.

1 plugin · aboyalejandro/eval-driven-development

An Agno agent that uses MCP tools and Skills to help with Substack content strategy. It connects to the Substack Author MCP server for data, uses Skills for domain expertise, tracks behavior with Opik, and can be exposed as an MCP server itself via AgentOS.

1 plugin · aboyalejandro/substack-author-agent