菜单
First-time setup for restruct. Runs rule generation (CLAUDE.md), verify wiring, permission defaults, and — if the feature flag is on — prompt refinement setup.
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
Disable restruct prompt refinement. Prompts pass through to Claude without local LLM preprocessing.
Enable restruct prompt refinement. Re-enables the local LLM preprocessing of prompts.
Generate a minimal, research-backed CLAUDE.md for this project. Analyzes the codebase for non-obvious commands and conventions, interviews you for team knowledge, and produces a concise file that stays under 30 rules.
Manually refine a prompt through the local LLM meta-prompt pipeline. Transforms a casual developer request into a structured, rules-aware prompt using a local Ollama model. Only works when features.refinement is enabled in config.yaml.
Review tool permission decisions across sessions and propose auto-approval rules. Analyzes which tools the user consistently approves, then suggests additions to .restruct/permissions.yaml to eliminate repetitive permission dialogs.
Write an initial .restruct/permissions.yaml with safe defaults so the restruct permit hook can auto-approve common tool use without repeated dialogs.
| name | setup-restruct |
| description | First-time setup for restruct. Runs rule generation (CLAUDE.md), verify wiring, permission defaults, and — if the feature flag is on — prompt refinement setup. |
This is the master setup flow for restruct. It orchestrates per-feature sub-skills
so you can configure the plugin end-to-end in one pass. Each sub-skill is also
directly invokable (e.g. /restruct:setup-verify-restruct) so you can re-run a single phase later.
Execute the steps in order. At each step, invoke the named sub-skill and wait for it to complete before moving on. Report progress briefly between steps.
Invoke the setup-rules-restruct skill. It generates or audits the project's CLAUDE.md file.
If the user declines or CLAUDE.md already exists and they are happy with it, continue.
Invoke the setup-verify-restruct skill. It discovers lint/typecheck/test/build commands
and writes .restruct/verify.yaml so Claude Code automatically runs them on task completion.
Invoke the setup-permissions-restruct skill. It writes an initial .restruct/permissions.yaml
with safe defaults for the detected ecosystem.
Check whether the refinement feature is enabled:
${CLAUDE_PLUGIN_ROOT}/bin/restruct config get features.refinement 2>/dev/null
If the value is true: invoke the setup-refine-restruct skill to install Ollama, pull a model, and enable refinement.
If the value is not true: print the following and skip to the summary:
"Prompt refinement is feature-flagged off in this release. Skipping Ollama / model setup. To opt in, set
features.refinement: truein~/.config/restruct/config.yamland run/restruct:setup-refine-restruct."
Print a summary of what was configured:
.restruct/verify.yaml checks (count and names).restruct/permissions.yaml status (created / skipped)Point the user at the plugin README (plugin/README.md in the repo, or the installed
plugin directory) for further configuration options.