Command Claude Code
You are the Cub Router. The user has typed /cub followed by a natural language request. Your job is to understand their intent and execute the right cub command or invoke the right cub skill.
Command Claude Code
You are the Architect Agent. Your role is to translate product requirements into a technical design that balances the project's needs with pragmatic engineering decisions.
Command Claude Code
You are running cub's code health audit. This checks for dead code, documentation gaps, and coverage issues.
Command Claude Code
You are the Capture Agent. Your role is to help someone think through a nascent idea and save it for later.
Command Claude Code
You are running cub's diagnostic tool to check configuration, hooks, and project health.
Command Claude Code
You are the Itemizer Agent. Your role is to break down the architecture into executable tasks that an AI coding agent (or human) can pick up and complete.
Command Claude Code
You are querying the task completion ledger. Run the appropriate subcommand based on the user's input.
Command Claude Code
You are the Orient Agent. Your role is to ensure product clarity before technical work begins.
Command Claude Code
You are launching cub's autonomous task execution loop. Determine the right flags from the user's input.
Command Claude Code
You are a Planning Agent. Your job is to read a feature specification and generate a set of beads-compatible issues (epics and tasks) for implementation.
Command Claude Code
You are the Spec Agent. Your role is to help someone articulate a feature idea and produce a well-structured specification file in specs/researching/.
Command Claude Code
You are executing the cub stage command, which bridges planning and execution by importing tasks from a completed plan into the task backend (beads or JSON).
Command Claude Code
You are showing project status. Run the appropriate status command based on the user's input.
Command Claude Code
You are getting smart suggestions for what to work on next, based on project state.
Command Claude Code
You are executing a cub task command. Run the appropriate subcommand based on the user's input and present the results.
Hook Claude Code
Runs after a tool call finishes, executing cub-hook.sh. From lavallee/cub.
Hook Claude Code
Runs when a session starts, executing cub-hook.sh. From lavallee/cub.
Hook Claude Code
Runs when the agent finishes a response, executing cub-hook.sh. From lavallee/cub.
Hook Claude Code
Runs when a session ends, executing cub-hook.sh. From lavallee/cub.
Hook Claude Code
Runs before the context is compacted, executing cub-hook.sh. From lavallee/cub.
Hook Claude Code
Runs when you submit a prompt, before the agent sees it, executing cub-hook.sh. From lavallee/cub.
Settings file Claude Code
Agent settings declaring 6 hook events (PostToolUse, SessionStart, Stop, SessionEnd).
Instructions file CodexOpenCode
Instructions for lavallee/cub, a project described as: A CLI for Ralph + Beads-style development with harnesses like Claude Code and Codex.
Instructions file
Instructions for lavallee/cub, a project described as: A CLI for Ralph + Beads-style development with harnesses like Claude Code and Codex.