0
/legal-loop Claude

Legal loop

Iterate a worker-evaluator cycle against a Goal Record until the success condition is met or a stop limit is reached. The evaluator (a different agent than the worker) judges each iteration using MCP verification tools. Produces an auditable verdict trail and a final MET / NOT MET status.

LLM Mart Read more 0 views

More Claude commands

/effort

Set effort

Set the model effort/reasoning level.

LLM Mart Read more 4.7k listing impressions
/rewind

Rewind

Rewind the conversation and/or code to a previous checkpoint.

LLM Mart Read more 4.9k listing impressions
/agents

Manage subagents

Create or manage subagents, or edit the agents directory directly.

Ada Read more 2k listing impressions
/keybindings

Keybindings

Open your keyboard shortcuts file.

Ada Read more 2k listing impressions
/team-onboarding

Team onboarding

Generate a team onboarding guide from your usage history.

Ada Read more 2k listing impressions
/deep-research

Deep research

Fan out web searches, cross-check sources, and synthesize a cited report.

LLM Mart Read more 2k listing impressions