一键导入
new-user
Welcome new DeepWork users on Pi and introduce native Pi workflows
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
Welcome new DeepWork users on Pi and introduce native Pi workflows
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。
Record a workflow by doing it, then turn it into a DeepWork job
Start or continue DeepWork workflows using native Pi tools
Set up DeepWork Reviews using .deepreview config files
Start structured planning that produces an executable DeepWork plan
Reference documentation for DeepWork Reviews
Create and manage DeepSchemas with validation and review generation
基于 SOC 职业分类
| name | new-user |
| description | Welcome new DeepWork users on Pi and introduce native Pi workflows |
| disable-model-invocation | true |
Guide a new user through DeepWork on Pi. This native Pi package uses Pi commands and tools directly; it does not use MCP.
Use deepwork_get_workflows to verify that DeepWork workflows are available in the current project. If the tool is unavailable, tell the user that the native deepwork package needs to be installed or reloaded, then suggest /reload or reinstalling the package.
If gh is installed, ask whether the user wants to star the repository. If yes:
gh api -X PUT /user/starred/ai-outfitter/deepwork
Skip this entirely if gh is unavailable.
Print a concise welcome message. Lead with: DeepWork makes AI agents reliable by giving them workflows, schemas, and reviews that verify process and output.
Cover:
.deepreview rules run by /review.If the current directory looks like a code project, ask whether the user wants automated review rules. If yes, use the configure-reviews skill.
Ask whether the user wants to record a workflow now. If yes, use the record skill. If no, tell them they can run /record anytime and /deepwork is the main workflow entry point.
Be brief, helpful, and concrete. Avoid long setup explanations unless a tool is missing or the user asks for details.