configurelisted

# Configure the project FAW works without this configuration. What changes is how much it has to assume. With no profile, the method resolves every missing value by the strictest option: it assumes there is a single environment and that it is production, so every write to the platform requires authorization in writing before it runs. That is correct as a default -- being wrong on the strict side costs one extra authorization, being wrong on the other side writes to production without asking -- but it can be more ceremony than the project needs. This skill asks the questions once and writes `.faw/config.json` in the working folder. ## Look before asking Several answers can be inferred from the repository rather than asked. Infer, propose, and confirm; do not ask what is already visible: - The git remote says whether the project lives on GitHub or Azure DevOps, which is a hint about where the tickets are. - The MCP servers connected in the session say which trackers can be operated directly. - The repository structure usually shows whether platform artifacts are versioned. ## The questions **1. Where do the tickets live?** Options: `ado`, `jira`, `github`, `internal`, `none`. If the user uses no tracking tool, or does not want to connect one, the answer is `internal`: FAW keeps the registry under `docs/faw/tickets/`, generates the identifiers, and git provides the history. Nothing has to be installed and no identifiers have to be invented. Whether an MCP server e