| name | dotfiles-bootstrap |
| description | Bootstrap a workstation with the dotfiles framework. Takes a GitHub
user / owner+repo / explicit clone URL and runs `dot init` (which
shells out to chezmoi) with the right safety prompts. Honors the
active agent profile (`ask` / `plan` / `apply` / `audit`) so it
defaults to dry-run in safer modes and full apply in `apply`.
|
| version | 1.0.0 |
| maintainer | Sebastien Rousseau <sebastian.rousseau@gmail.com> |
| license | MIT |
| tags | ["dotfiles","chezmoi","bootstrap","provisioning","workstation","agents"] |
| requires | ["chezmoi >= 2.47","bash >= 4.0","git"] |
| homepage | https://github.com/sebastienrousseau/dotfiles |
| docs | https://doc.dotfiles.io/manual/ |
dotfiles-bootstrap
Provision a new workstation from a public dotfiles repository through
the sebastienrousseau/dotfiles harness.
This skill is for Claude Code to invoke dot init <user> on the user's behalf — most useful when the user says "set me up like Alice" or "bootstrap a fresh laptop using my GitHub dotfiles."
What this skill does
When invoked, it:
- Verifies the target argument is one of:
- a GitHub username (e.g.
alice → expands to github.com/alice/dotfiles)
- an
owner/repo pair (e.g. alice/configs)
- a full HTTPS or
git@host:path clone URL
- Refuses plain
http:// URLs (no transit integrity).
- Reads the user's current agent profile from
$DOT_AGENT_PROFILE (defaults to ask):
ask → run dot init <user> --dry-run (read-only preview)plan → run dot init <user> --dry-run and explain the planapply → run dot init <user> (full apply with chezmoi apply)audit → run dot init <user> --no-apply (clone only, no chezmoi apply)
- Surfaces the trust prompt warnings before any clone runs — the target repo's scripts execute with the user's privileges.
- Reports the outcome (success / failed / aborted-by-trust-prompt) so the user can decide next steps.
Invocation pattern
dot init <user>
dot init <owner>/<repo>
dot init <https-or-ssh-url>
dot init <user> --dry-run
dot init <user> --no-apply
dot init <user> --force
Profile-aware defaults
| Profile | Default flag set | Notes |
|---|
ask | --dry-run | Show the plan; ask the user before running for real. |
plan | --dry-run | Same plan, augmented with chezmoi diff if available. |
apply | (no flags) | Full apply; interactive trust prompt remains. |
audit | --no-apply | Clone the source; let the user inspect before apply. |
Failure modes the skill surfaces
chezmoi not installed → instruct user to run install.sh first- URL refuses HTTPS check → ask user for explicit confirmation if SSH
- Target chezmoi source dir already exists → ask user about
--force
- Trust prompt aborted → return cleanly, no state mutation
chezmoi init non-zero exit → surface the chezmoi stderr to the user
Why this skill exists
Closes the round-1 audit's adoption gap "bootstrap a foreign dotfiles repo through this framework's harness." Without the skill, a Claude Code session has to remember the right flag set for the user's current trust posture; with the skill, the profile dictates the safe default.
References
