kbd-init

To initialize project configuration, run the orchestrator setup command from the repository root to auto-discover project identity from context files like AGENTS.md, README.md, and package.json, generating the required orchestrator configuration.

The project setup inference reads repository context files including AGENTS.md, CLAUDE.md, README.md, package.json, Cargo.toml, and pyproject.toml to extract the project name, technology stack, spec paths, and workspace planning details.

No, the initialization process keeps the skill directory read-only from the project perspective, generating project-scoped configuration files in the .kbd-orchestrator directory without modifying existing skill directories.

Yes, you can generate project constraints and context files for Python or Rust projects by auto-discovering configurations from pyproject.toml or Cargo.toml to populate the generated orchestrator files with the correct technology stack.

You need to run the orchestrator bootstrap command once per project before executing any other orchestrator commands, ensuring the .kbd-orchestrator/project.json and constraints.md files are generated from your repository context.

The project initialization process generates two files: .kbd-orchestrator/project.json containing project context, stack commands, and tool preferences, and .kbd-orchestrator/constraints.md containing project-specific blocking and warning rules.