Initialize project-level Claude setup (gitignore entries, settings.local.json) without copying agent/skill files. Use this in projects where agents were installed globally (default setup.sh behavior).
The English summary is being prepared. Showing the original description.
Review the instructions the agent will receive before you install.
You are initializing project-level Claude configuration for this repository. Follow the steps below strictly in order.
Check whether .claude/agents/ or .claude/settings.local.json exists in the current working directory.
.claude/agents/ exists: inform the user that this project already has a full local Claude install and no initialization is needed. Stop here..claude/settings.local.json exists: inform the user that this project already has project-level Claude setup and no initialization is needed. Stop here..gitignore in the project root. If it does not exist, treat it as empty.# .claude section comment if you are adding any:# .claude
.devcontainer/.env
.claude/settings.local.json
.claude/auth.local.md
.claude-worktrees/
.DS_Store
tasks/**
qa-output/**
debug-output/
.playwright-cli/
# .claude section)..claude/settings.local.json exists.{
"permissions": {
"allow": [],
"deny": []
}
}
Print a clear summary of what was done:
## Claude project setup complete
- .gitignore: [added N entries / already up to date]
- .claude/settings.local.json: [created / already existed]
### Next steps
- Run `claude login` if you haven't already
- Edit `.claude/settings.local.json` to configure per-project tool permissions
- Agents and skills are available globally via your ~/.claude/ install
- The /qa skill uses `playwright-cli` for browser automation — run `playwright-cli install` to set it up
No reviews yet. Tell others what you thought.
No login needed. Share how it worked for you, ask a question, or note where it did not run.
No comments yet. Be the first to write one.
Lightweight build: explore → plan → approve → implement in a single Opus context, no sub-agent fan-out. Verifies build + tests, optionally commits, and offers a separate /code-review pass at the end. Use by default for everyday features; reach for /claude-setup:build only when the work needs parallel fan-out or exceeds one context.
The English summary is being prepared. Showing the original description.
Lightweight build: explore → plan → approve → implement in a single Opus context, no sub-agent fan-out. Verifies build + tests, optionally commits, and offers a separate /code-review pass at the end. Use by default for everyday features; reach for /build only when the work needs parallel fan-out or exceeds one context.
The English summary is being prepared. Showing the original description.
Reads task files from tasks/<branch>/ and the diff against the detected base ref, then generates a polished PR description in markdown for the user to copy-paste.
The English summary is being prepared. Showing the original description.
Debug pipeline: investigates a bug, diagnoses root cause, writes failing tests, implements fix via TDD, and reviews the result. Orchestrates bug-investigator -> bug-fixer -> code-reviewer.
The English summary is being prepared. Showing the original description.
Interview the user relentlessly about a plan or design until reaching shared understanding, resolving each branch of the decision tree. Use when the user wants to stress-test a plan, get grilled on their design, or mentions "grill me". Run BEFORE /claude-setup:build when a feature idea is fuzzy and a PRD doesn't yet exist.
The English summary is being prepared. Showing the original description.
Interview the user relentlessly about a plan or design until reaching shared understanding, resolving each branch of the decision tree. Use when the user wants to stress-test a plan, get grilled on their design, or mentions "grill me". Run BEFORE /build when a feature idea is fuzzy and a PRD doesn't yet exist.
The English summary is being prepared. Showing the original description.