Agent repo init
One-click initialization of a multi-agent repository from the RepoBrain template. Use this skill when users want to scaffold a new project quickly (`quick` mode) or with runtime defaults (`full` mode) including MCP toggle, swarm preference context, sandbox type, and optional git init. LLM configuration is handled later by rb-setup.From its SKILL.md
npx -y skills add study8677/repobrain --skill agent-repo-initAssembled from the repository path, not quoted from the project. Check it against their README if it does not work.
SKILL.md
1.6 KB, as published. Nobody here has run it
Agent Repo Init
Initialize a new project from this repository template with two modes.
Modes
quick: Fast scaffold with clean copy and minimal setup.full:quickplus runtime profile setup (.env, mission, context profile, init report) and optionalgit init.
Run via Script
Use the portable script in this skill directory:
python skills/agent-repo-init/scripts/init_project.py \
--project-name my-agent \
--destination-root /absolute/path \
--mode quick
Full mode example:
python skills/agent-repo-init/scripts/init_project.py \
--project-name my-agent \
--destination-root /absolute/path \
--mode full \
--enable-mcp \
--disable-swarm \
--sandbox-runtime microsandbox \
--init-git
Expected Output
- New project at
<destination_root>/<project_name> - Clean copy without local runtime state
- Initialization report at
artifacts/logs/agent_repo_init_report.md - Script is self-contained and does not import project
src/modules
Notes
- Keep destination outside the current template repository.
- For
fullmode, review.context/agent_runtime_profile.mdafter generation. - Run
/rb-setupin the generated project to choose and write the LLM endpoint.