大师研究笔记工作流
PassAudited by ClawScan on May 4, 2026.
Overview
This is a coherent research-and-Obsidian note-taking workflow, with some ordinary privacy and persistence considerations but no evidence of hidden or malicious behavior.
This skill appears safe to use for its stated purpose. Before using it, choose a narrow Obsidian vault path, avoid sharing unnecessary personal or business details with delegated research tasks, review the Phase 1.5 quality summary, and check the final Markdown note before treating it as part of your trusted knowledge base.
Findings (3)
Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.
The agent may inspect filenames or matching entries in your Obsidian vault to avoid duplicate notes.
The skill documents a local shell command to inspect the Obsidian vault for related notes. This is purpose-aligned and scoped to the user's vault, but it is still local filesystem access.
ls /path/to/vault/ | grep -i "关键词"
Use a specific vault path, avoid pointing it at unrelated private directories, and review what the agent found before allowing it to write new notes.
Personal details, business context, or excerpts from existing notes could be included in delegated research prompts.
The workflow instructs the agent to run multiple delegated research tasks and provide them with user context and related note context. This supports personalized synthesis, but it can share sensitive personal or business context across subtasks.
启动 **6 个 delegate_task** 并行搜索... 需要提供的 context 需包含 - 用户基本情况（职业、地点、已有布局） - 用户已有的相关笔记或认知基础
Provide only the minimum context needed, remove sensitive details before delegation, and confirm that delegated tasks are allowed to see the information you provide.
Incorrect or biased research could become part of your long-term knowledge base if not reviewed.
The workflow writes synthesized AI research into a persistent Obsidian Markdown file. Persistent notes may later be reused as context, so source quality and limitations matter.
文件路径：`/path/to/vault/[名字] 核心工具箱.md`
Keep the required references and 'honest boundaries' section, review the note before relying on it, and treat AI-generated conclusions as research aids rather than verified facts.