Local Bookmark Librarian
v1.0.0去重和再分类本地导出的书签或链接清单，生成主题索引和维护建议。；use for bookmarks, links, knowledge workflows；do not use for 直接修改浏览器配置, 删除用户未确认链接.
⭐ 0· 97·0 current·0 all-time
byvx：17605205782@52yuanchangxing
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
OpenClaw
Benign
medium confidencePurpose & Capability
Name/description (de-duplicate and re-categorize local bookmark/link exports) matches the included resources: SKILL.md, resources/spec.json, template, examples, and a Python script that produces structured reports and directory/csv audits. Required binaries are limited to python3, which is appropriate.
Instruction Scope
SKILL.md explicitly limits actions (read-only, produce drafts, do not modify browser config or delete links) and directs use of the bundled script; however the script can audit arbitrary directories or files if the user points it at them (it deliberately scans many text/code suffixes). That broad file scanning is consistent with a 'directory_audit' mode, but users should avoid running it against whole home or system directories unless they intend an audit (it may surface sensitive items from other projects).
Install Mechanism
No install spec — instruction-only with an included Python script. This is low-risk: no network installs, no archives downloaded, and the script is local and uses only the standard library.
Credentials
The skill declares no required environment variables or credentials and the code does not attempt to read secrets or external config. No disproportionate credential or config access is requested.
Persistence & Privilege
always:false and user-invocable. The skill does not request permanent presence or modify other skills or system-wide settings. The script can write an output file (user-specified) but offers a --dry-run mode; this matches the stated read-only/audit-first approach.
Assessment
This skill appears coherent and designed to run locally under python3 to analyze bookmark exports or directories and emit a reviewable Markdown report. Before running: (1) run it on a copy/export of your bookmarks (as the README suggests) rather than directly on a live browser profile; (2) avoid pointing the script at broad directories like your entire home or code repositories unless you intend an audit (it will scan .py/.js/.sh/.md, etc., and may surface sensitive strings); (3) review generated output before performing any write/delete/publish actions — the skill emphasizes dry-run and review-first workflows; (4) the included script runs local code, so only execute it if you trust the skill source. If you want a higher assurance level, request the full, non-truncated script review or run the script in a sandboxed environment first.Like a lobster shell, security has layers — review code before you run it.
latestvk97d12aq72475zmyrpa6480p5s8341sq
License
MIT-0
Free to use, modify, and redistribute. No attribution required.
Runtime requirements
🔖 Clawdis
OSmacOS · Linux · Windows
Binspython3