Minimal Agent

Where It Fits

When building an agent stack, many teams first need a small, identifiable, invokable skill placeholder to validate routing, permissions, prompt assembly, and response shape before adding complex tooling. Minimal Agent is classified under AI Agent, and the provided metadata is sparse: it does not list a model, tools, database, or business workflow. It is therefore better read as a lightweight skill entry point than as a complete production agent.

How To Reason About It

Under the general skill-site pattern, the skill appears under @user_e02e04b8/minimal-agent. A host agent would typically:
- declare or load the skill;
- provide task, context, and expected output in the prompt;
- let the host platform decide whether it is exposed as a tool, sub-agent, or prompt template;
- inspect the returned content before taking next steps.

Because SKILL.md does not define additional capability fields, the observed behavior mainly depends on the calling context and host-platform conventions. This entry is useful for validating a minimal agent skill identity, source, and invocation boundary. It should not be assumed to include reasoning chains, tool execution, memory, or multi-turn orchestration without more documentation.