Claude Skills

Agent Grade and Iterate Loop

Phase 3 of building a the agent Managed Agent — the bounded grade→iterate loop.

#agents#claude skill#source: alirezarezvani/claude-skills (MIT)

Preview

This is the plugin's loop: CMA's outcome primitive self-grades the agent's work in an isolated context and feeds failing verdicts back for the next attempt.…

Full prompt is for subscribers

Get instant access to this and every other prompt in the library.

Unlock for $14.89/month

You're seeing the public preview

Subscribe to unlock the full prompt, recommended model, and usage notes for every entry in the library.

Get full access — $14.89/month