Skip to content
Discussion options

You must be logged in to vote

The clean mental model is:

  • commands = user entrypoints / workflows
  • agents = specialist roles / execution styles

A command does not need to literally embed an agent filename to be using the agent pattern. Sometimes the command is the operator-facing workflow, and the agent is the specialist you delegate to once that workflow starts.

So in practice:

  • /refactor-clean is the workflow prompt you run
  • refactor-cleaner is the specialist role that can carry that workflow out

The repo is not perfectly 1:1 command-to-agent. Some commands are direct workflows, some are thin wrappers around agent behavior, and some can be run either way. I agree the mapping should be documented more explicitly.

Replies: 1 comment

Comment options

You must be logged in to vote
0 replies
Answer selected by affaan-m
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Category
Q&A
Labels
None yet
2 participants