Personal API
ReviewAudited by ClawScan on May 8, 2026.
Overview
This appears to be a coherent Obsidian-vault scaffolding skill, but it intentionally gives AI assistants persistent personal context and local note-management instructions, so users should review privacy settings before use.
This skill looks safe for its intended local Obsidian-vault purpose. Before installing, review the setup script, confirm the vault path, avoid putting secrets in ME.md or AGENT.md, add your own .gitignore/version-control exclusions, and review any AI-created or AI-moved notes before relying on them.
Findings (4)
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.
Running the setup script will create files and folders in the Obsidian vault path you provide.
The skill includes a user-run shell script that creates directories and copies templates into the selected vault. This is expected for a scaffolding skill and is quoted/validated, but it is still local code execution.
Usage: export OBSIDIAN_VAULT_PATH="/path/to/your/vault"; bash scripts/setup.sh [--minimal] ... mkdir -p "$VAULT/00.context/projects/active" ... cp "$src" "$dst"
Review the script first, confirm OBSIDIAN_VAULT_PATH points to the intended vault, and use --minimal if you only want the identity files.
An AI assistant following these instructions may create, move, or reorganize knowledge notes in your vault.
The installed methodology tells future AI assistants they may create and reorganize files under the knowledge area, while prohibiting bulk deletion without explicit confirmation. This is purpose-aligned but gives agents local note-management authority.
| 在 `30.knowledge/` 下新建文件 | ✅ | ... | 整理、迁移 `30.knowledge/` 下的文件 | ✅ | ... | 批量删除文件 | ❌（需要用户明确确认） |
Keep backups or version history for the vault, and require explicit approval before large reorganizations or any deletion.
Personal details placed in ME.md, AGENT.md, or linked vault notes may be read and used by AI assistants that have access to the vault.
The skill is designed to make personal identity and knowledge-base content available as persistent AI context. This is the core feature, but it means stale, sensitive, or misleading notes can affect future agent behavior.
整个 vault 就是我的"第二大脑"，是为了让你能**像另一个我**一样理解我而设计的。
Do not put secrets or highly sensitive information in these files, review them periodically, and restrict vault access for agents that do not need it.
Users might assume version-control protection exists when they may need to create it themselves.
The provided file manifest does not include a .gitignore file, so the privacy-helper claim is not supported by the supplied artifacts. The skill does still warn users not to commit personal files.
A `.gitignore` is included to help. This skill ships only templates — never your data.
Manually add ME.md, AGENT.md, and sensitive vault paths to .gitignore or another exclusion mechanism before using a public or synced repository.