一键导入
setup
Configure next-level for your project — detects languages, installs linters/formatters, writes config
用 Codex 或 Claude 帮你安装 复制这段 Prompt,粘贴到 Codex、Claude 或其他助手里,让它检查 Skill 页面并帮你完成安装。
菜单
Configure next-level for your project — detects languages, installs linters/formatters, writes config
用 Codex 或 Claude 帮你安装 复制这段 Prompt,粘贴到 Codex、Claude 或其他助手里,让它检查 Skill 页面并帮你完成安装。
基于 SOC 职业分类
| name | setup |
| description | Configure next-level for your project — detects languages, installs linters/formatters, writes config |
| user_invocable | true |
You are running the next-level setup wizard. Your job is to detect the project's languages, check for required tooling, install what's missing, and write the configuration file.
Run the dependency detection engine to analyze the current project:
python3 -c "
import sys
sys.path.insert(0, '${CLAUDE_PLUGIN_ROOT}/lib')
import json
from dependencies import full_dependency_check
result = full_dependency_check('$(pwd)')
print(json.dumps(result, indent=2))
"
Parse the output. Report to the user:
For each missing tool, ask the user if they want to install it. Group by language.
Install commands by language:
npm install -g prettier eslintuv tool install ruff and uv tool install basedpyrightbrew install swiftformat swiftlintrustup component add rustfmt clippy rust-analyzergo install github.com/golangci/golangci-lint/cmd/golangci-lint@latest and go install golang.org/x/tools/gopls@latestRun each install command the user approves. If an install fails, report the error but continue with remaining tools.
If any LSP servers are available, write .lsp.json to the project root:
{
"lspServers": {
"basedpyright": { "command": "basedpyright-langserver", "args": ["--stdio"] },
"vtsls": { "command": "vtsls", "args": ["--stdio"] },
"sourcekit-lsp": { "command": "sourcekit-lsp" },
"rust-analyzer": { "command": "rust-analyzer" },
"gopls": { "command": "gopls", "args": ["serve"] }
}
}
Only include entries for LSP servers that are actually installed. Skip this step if no LSP servers are available.
Write ~/.next-level/config.json using the Python config module:
python3 -c "
import sys
sys.path.insert(0, '${CLAUDE_PLUGIN_ROOT}/lib')
import json
from config import write
config = {
'setup_complete': True,
'project_root': '$(pwd)',
'languages_detected': $LANGUAGES_JSON,
'features_enabled': {
'file_checker': True,
'comment_stripping': True,
'tdd_enforcement': True,
},
'plugins_available': $PLUGINS_JSON,
'linters': $LINTERS_JSON,
}
write(config)
print('Config written successfully')
"
Replace the $VARIABLES with actual JSON values from the detection results.
The linters field should map each language to its available tools by role:
{
"python": {"formatter": "ruff", "linter": "ruff", "type_checker": "basedpyright"},
"typescript": {"formatter": "prettier", "linter": "eslint"}
}
Only include tools that are actually installed.
Show the user a summary:
next-level setup complete!
Languages: python, typescript
Features: file_checker, comment_stripping, tdd_enforcement
Linters:
python: ruff (format+lint), basedpyright (types)
typescript: prettier (format), eslint (lint)
Plugins:
omega-memory: not installed (optional — enhanced memory across sessions)
coderabbit: not installed (optional — AI code review in /spec-verify)
Config: ~/.next-level/config.json
LSP: .lsp.json (written to project root)
Run /next-level:doctor anytime to check health.
jq is not installed, warn the user (needed for bash hook scripts)Primary review entrypoint for plans, diffs, commits, subagent results, and pre-submit checks. Selects quick, standard, or deep mode by scale and risk.
Planner-led repo improvement loops with subagents, review gates, Mermaid DAGs, and workflow-aware commits. Use for improve, refactor, audit, or continue-loop requests.
Triage and apply code review feedback with technical verification. Use when a human, reviewer subagent, CI, or adversarial reviewer returns findings or requested changes.
Heavy review mode normally invoked by request-code-review for risky checkpoints, pre-submit stacks, OpenSpec/cflx acceptance, security-sensitive work, or explicit deep review, audit, or red-team requests.
Post-implementation cleanup for recent changes using parallel reuse, quality, and efficiency reviewers. Use after building a feature or fix, before commit, PR, or submit.
End-of-workflow learning loop for skill-guided work. Use at session/checkpoint closeout to classify GitHub/CodeRabbit/human/CI findings, log missed-review patterns, and propose skill improvements only when evidence justifies them.