agentsclimarketplace

Repo init

Skill RadOrigin-LLC/RAD-Claude-Skills/plugins/rad-repo-manager/skills/repo-init

Marketplace of plugins and skills for Claude Code

Install
npx -y skills add RadOrigin-LLC/RAD-Claude-Skills --skill repo-init

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

One thing to look at

  • 5 stars5 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

What its author says it does

Copied from the file, not written here

This skill should be used when the user says "set up this repo", "bootstrap this project", "new project setup", "scaffold the docs", "initialize the repo model", "get me started right", or when startup recommended it on a fresh repo. Builds the container for the doc model in a new or nearly empty repo: an AGENTS.md skeleton with the stamped doc-model block, the CLAUDE.md @AGENTS.md shim, docs/ + docs/archive/, and a handoff stub. Creates no shelf or initiative docs (their triggers haven't fired), runs no fit-out, never invents product content, and ends by recommending /rad-planner:plan. For an existing repo with history and docs, use /rad-repo-manager:adopt instead.

SKILL.md

5.3 KB, as published. Nobody here has run it

Repo Init — build the container

Lay the foundation for a fresh, nearly empty repo. Build the container only — the shelf docs (decisions.md, ideas.md, prd.md, plan.md, …) are created later, each when its trigger fires (see <plugin-root>/references/shelf-spec.md). Do not write product, plan, or design content — that's the owner's and /rad-planner:plan's job. Templates live in <plugin-root>/templates/ (where <plugin-root> is either ${CLAUDE_PLUGIN_ROOT}, global config plugins/rad-repo-manager, or local workspace plugins/rad-repo-manager).

Use this only for greenfield setup. An established repo with history and docs goes to /rad-repo-manager:adopt (archaeology first); an on-model repo with drift goes to /rad-repo-manager:repo-align.

1. Look before scaffolding

git status --short
git remote get-url origin   # for the AGENTS.md remote line, if set

Glob the repo. Detect the app type and package manager if obvious (package.json, lockfiles, pyproject.toml, Cargo.toml, framework config) — this only informs the AGENTS.md identity line, not product content.

Confirm this is greenfield: no AGENTS.md, no real docs, little or no git history. If the repo is established, stop and recommend adopt.

2. Agent scope

If --agents was passed, use it. Otherwise ask (via AskUserQuestion or ask_question on Antigravity): which coding agents will use this repo — Claude, Codex, Gemini, or a mix? This decides which shims to create. Codex reads AGENTS.md natively (no shim). If unclear, ask before creating non-Codex shims.

3. Scaffold — container only, only what's missing

Create only files that don't already exist. Never overwrite a user-authored file without explicit confirmation.

  • AGENTS.mdtemplates/AGENTS.md — the L0 skeleton: identity line, deploy: line, stack/commands placeholders, the doc-model block (stamp it verbatim from templates/doc-model-block.md — it's what makes every other tool conform), the "How we work" contract, and the hard-rule slots. Keep it within its 40-line budget.
  • CLAUDE.mdtemplates/CLAUDE.md — the one-line @AGENTS.md shim.
  • GEMINI.mdtemplates/GEMINI.md — only if Gemini is in scope.
  • docs/ and docs/archive/ folders; docs/archive/README.mdtemplates/archive-README.md. docs/initiatives/ is created only when an approved migration uses templates/initiative.md; do not scaffold an empty initiative.
  • docs/handoff.mdtemplates/handoff.md — the stub, placeholder values, including the ## Deferred — do not re-raise section.

Do not create any shelf doc — no prd.md, plan.md, decisions.md, ideas.md, lessons.md, design.md, architecture.md, or api.md. Their triggers haven't fired; /rad-planner:plan births prd/plan, and the rest appear on first real content. Do not run fit-out here (no code yet means no traits to detect — that happens at the first /rad-repo-manager:ship or during adopt). Do not create docs/status.md, docs/roadmap.md, docs/inbox/, loose root-level status docs, or speculative scoped instructions. A scoped AGENTS.md plus one-line CLAUDE.md shim is allowed later only when code provides evidence of materially different subtree rules.

4. Fill only what's mechanical

Set the remote line from git remote get-url origin if available. Leave deploy: as none unless the user names a target. Stamp the handoff stub's **Updated:** with today's date. Note in one line that the plugin's hooks (SessionStart banner, Stop drift nudge, PreCompact checkpoint) are universal — they activate on their own now that AGENTS.md + docs/handoff.md exist; nothing to install per-repo. Leave every other <PLACEHOLDER> for the owner. Do not invent goals, stack, rules, or validation promises.

5. Hand off

Tell the user, plainly: the container is in place; the docs themselves get created when they're first needed. The next step is /rad-planner:plan — its interview births docs/prd.md and docs/plan.md and seeds the first decisions and ideas.

Output format

Repo init:
Detected repo type:
Agent scope:
Files created:
Files skipped (already exist):
Next step for the user:   /rad-planner:plan
Notes:

References

  • <plugin-root>/templates/ — AGENTS.md skeleton, doc-model block, shims, handoff stub, initiative, archive README, repository config
  • <plugin-root>/references/shelf-spec.md — the shelf, triggers, budgets, one-writer table

Keep looking

Skills are one crate of 328,083. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.